TW200934250A - Improved enhancement layer coding for scalable video coding - Google Patents

Abstract

This disclosure describes scalable video coding techniques. In particular, the techniques may be used to encode refinements of a video block for enhancement layer bit streams in a single coding pass, thereby reducing coding complexity, coding delay and memory requirements. In some instances, the techniques encode each nonzero coefficient of a coefficient vector of the enhancement layer without knowledge of any subsequent coefficients. Coding the enhancement layer in a single pass may eliminate the need to perform a first pass to analyze the coefficient vector and a second pass for coding the coefficient vector based on the analysis.

Description

200934250 IX. INSTRUCTIONS: and more particularly in relation to the field of view [Technical Field of the Invention] The present invention relates to an adjustable video writing code for digital video writing code data. The present application claims the benefit of U.S. Provisional Application No. 60/979,919, filed on Jan. 15, 2007, and U.S. Provisional Application Serial No. 60/940,214, filed on Jan. 16, 2007. The content of each of these applications is incorporated herein by reference.

[Prior Art] Digital video capabilities can be combined with a wide range of devices, including digital televisions, digital live broadcast systems, wireless communication devices, wireless broadcast systems, personal digital assistants (PDAs), laptops or desktop computers. Digital cameras, digital recording devices, video game devices, video game consoles, cellular or satellite radiotelephones and the like. Digital video equipment implements video compression technology such as Motion Picture Experts Group (MP Call 2, MpEG_4 or International Telecommunication Union Standardization Sector (ITU_T) H.264/MPEG_4 Part 1 Advanced Video Write Code (AVC) (hereafter ordered as &quot ; H 264 / MpEG 4 part i 〇 A% standard) to transmit and receive digital video more efficiently. Video M technology performs spatial and temporal prediction to reduce or remove the inherent redundancy in the video sequence. Video compression typically includes spatial pre- and/or motion estimation and motion compensation to generate predictive video blocks. In-frame write predictions are used to reduce or remove a given code unit (eg, an empty gate between blocks) ^ 圃I times film) The number of work in the room. For the video encoder test to compress the data based on other information in the same level of the syllabus - good ^ π work top code. , I35453.doc 200934250 Inter-frame coding relies on temporal prediction to reduce or remove temporal redundancy between video blocks of consecutive video frames of a video sequence. Therefore, for inter-frame writing, &, video encoder execution Motion estimation and exercise Compensating to track the movement of matching video blocks of two or more adjacent code units. After spatial or temporal prediction, by subtracting the predicted video block from the original video block being coded The block of residual coefficients (called the remaining block or residual information). The remaining block can be a two-dimensional matrix of the coefficient values of the difference between the predicted video block and the original block. The video encoder can be used for the remaining area The block applies transform, quantization, and entropy write processes to further reduce the bit rate associated with communication of the remaining blocks. The transform technique may include discrete cosine transform (DCT), wavelet transform, integer transform, or other type of transform. The 'for example' transform process converts the set of pixel domain coefficients into transform coefficients, which represent the energy of the pixel domain coefficients in the frequency domain or the transform domain. The quantization coefficients are applied to the transform coefficients to produce quantized transform coefficients. Limiting the number of bits associated with any given coefficient. The video encoder entropy encodes the quantized transform coefficients to further compress the quantized coefficient of variation The video encoder may entropy encode the coefficients using variable length write codes, arithmetic write codes, fixed length write codes, or a combination thereof. The video decoder may perform an inverse operation to reconstruct the video sequence. Some video writes such as MPEG-2 The code standard encodes video with relatively constant quality, bit rate or spatial resolution. This technique may be sufficient to provide video applications to devices with similar decoder capabilities (eg, memory or processing resources) and/or connection quality. However, most modern video transmission systems typically include devices with varying decoder capabilities and/or connection quality. In systems such as 135453.doc 200934250, video transmissions encoded with relatively constant quality, bit rate or spatial resolution are transmitted. Causes the video application to work for devices with the appropriate decoder capabilities and/or connection quality' and does not work for devices that do not have the proper decoder capabilities and/or connection quality. In wireless situations, for example, close to video transmission sources. Devices may have higher quality connections than devices that are located away from the source. Similarly, devices that are located away from the source may not be able to receive encoded video transmitted with strange quality, bit rate, or spatial resolution. Other video writing standards use adjustable code writing techniques to overcome these problems. (for example) a tunable video write code (SVC) according to an extension of ITU-T H.264/MPEG-4 Part 10 AVC, which refers to encoding a video sequence as a base layer and one or more adjustable enhancement layers. Video writing code. For SVC, the base layer typically carries video material with a basic space, time and/or quality level. One or more enhancement layers carry additional video material to support higher space times and/or quality levels. The enhancement layer can, for example, add spatial resolution to the frame of the base layer, or additional frames can be added to increase the total frame rate. In some examples, the base layer can be transmitted in a more reliable manner than the transmission of the enhancement layer. Similarly, a device that is remote from the encoded video source location or has a lower decoder capability may be able to receive the base layer and thus receive the video sequence even at the lowest space, time and/or quality level. SUMMARY OF THE INVENTION The present invention describes an adjustable video write code technique that allows entropy coding enhancement layer turbulence in a single write once operation. Conventionally, a plurality of write code operations are used to encode the enhancement layer bit stream. For each video block of the enhancement layer, for example, 'first write code one operation can collect statistics on the block used in the code code table (or codebook) selected for the entropy code writing area 135453.doc 200934250 block. The data, and the second write code can be used to extract the coded block using the selected write code table. However, the technique of the towel according to the present invention entropy encodes the video block of the enhancement layer bit stream in the case of the first line of the statistical data used in the video code table selection.

The reality is to encode the enhancement layer using a write-coding technique that encodes the coefficients of the enhancement layer on a coefficient-by-coefficient basis in a single-write code-second operation. In one example, for each of the non-zero coefficients of the enhancement layer video block, the video encoder can encode the end of block (EOB) symbol, the length of the run, and the sign. The video encoder can encode the video layer of the enhancement layer using only a single-write code table, thereby eliminating the need to perform a first write-and-batch operation of the statistics of the selected write code table. In addition, the video encoder may not encode the magnitude of the non-zero coefficients in the enhancement layer. In this way, the magnitude of all non-zero coefficients of the enhancement layer can be limited to a magnitude that does not encode the coefficient of the enhancement layer, which can result in a certain loss of peak-to-noise ratio (PSNR) but is reduced for coding enhancement. The number of bits of the layer, °, the techniques of the present invention may provide several advantages. For example, such techniques can reduce write code complexity, write code latency, and memory requirements for encoding enhancement layer bitstreams while maintaining write efficiency. Using an adjustable video write code to encode video data, I 3 uses the first quality coded video block as part of the base layer bit stream. The method also includes refining the encoded video block as part of at least one enhanced layer bit stream, the refinement of the video blocks causing greater than the first quality when combined with the video block encoded with the __ quality The second quality of the view 135453.doc 200934250 block. The method also includes encoding the refinement of the video block in a single coded one operation. In another aspect, an apparatus for encoding video data using an adjustable video write code includes at least one encoder that encodes a video block as a portion of a base layer bit stream with a first quality, And the refinement of the encoded video_block is part of at least one enhanced layer bitstream, and the refinement of the video blocks causes a second greater than the first quality when combined with the video block encoded with the first quality Quality video block. The refinement of the video block is encoded in a single coded operation. In another aspect, a computer readable medium containing instructions for causing one or more processors to encode a video block as a portion of a base layer bitstream with a first quality; and encoding the video block The refinement is a portion of the at least one enhancement layer bitstream. The refinement of the video blocks results in a video block having a second quality greater than the first quality when combined with the video block encoded with the first quality. The refinement of the video block is encoded in a single coded one operation. In still another aspect, an apparatus for encoding video data using an adjustable video write code includes: a first component for encoding a video block as a portion of a base layer bit stream with a first quality; and Refining the coded video block as a second component of at least one portion of the enhancement layer bitstream, the refinement of the video blocks causing greater than the first quality when combined with the video block encoded with the first quality The second quality video block. The refinement of the video block is encoded in a single coded operation. In another aspect, a method of decoding video data using an adjustable video write code includes decoding a base layer bitstream to obtain a first quality video zone 135453.doc • 10 - 200934250 block 'and decoding enhancement layer The metadata stream is refined to obtain a video block, and the refinement of the video blocks results in a second quality video block when combined with the first quality decoded video block. The decoding enhancement layer includes decoding, for each non-zero coefficient of the refinement of the video block, a symbol indicating the presence of at least one residual non-zero coefficient, a run length indicating the number of zero-valued coefficients before the non-zero coefficient, and a non-zero The sign of the coefficient. In yet another aspect, an apparatus for decoding video data using an adjustable video write code includes at least one decoder that decodes a base layer bitstream to obtain a first quality video block And decoding the enhancement layer turbulence to obtain refinement of the video block, and the refinement of the video blocks results in a second quality video block when combined with the first quality decoded video block. At least one decoder decodes, for each non-zero coefficient of the refinement of the video block, a symbol indicating the presence of at least one residual non-zero coefficient, a run length indicating a number of zero-valued coefficients before the non-zero coefficient, and a non-zero The sign of the coefficient. In still another aspect, a computer readable medium containing instructions for causing one or more processors to decode a base layer bitstream to obtain a first quality video block; and decoding the enhancement layer bitstream To achieve refinement of the video blocks, the refinement of the video blocks results in a second quality video block when combined with the first quality decoded video block. The instructions cause the one or more processors to decode a symbol indicating the presence of at least one residual non-zero coefficient and a number of zero-value coefficients prior to the non-zero coefficient for each non-zero coefficient of refinement of the video block The length and the sign of the non-zero coefficient. In another aspect, an apparatus for decoding a 135453.doc -11-200934250 data using an adjustable video write code includes: decoding a base layer bitstream to obtain a first quality video block a first component; and a second component for decoding the enhancement layer bitstream to obtain refinement of the video block, the refinement of the video blocks being combined with the first quality decoded video block to cause Two quality video blocks. The second decoding component decodes, for each non-zero coefficient of the refinement of the video block, a symbol indicating the presence of at least one residual non-zero coefficient, a run length indicating the number of zero-valued coefficients before the non-zero coefficient, and The sign of the non-zero coefficient. The technique described in the present invention can be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the software may be executed in a processor. The processor may refer to one or more processors, such as a microprocessor, an application specific integrated circuit (ASIC), a field programmable gate array ( FPGA) or digital signal processor (DSP), or other equivalent integrated or discrete logic circuits. Software containing instructions for performing such techniques may initially be stored in a computer readable medium' and loaded and executed by a processor. Thus, the present invention also contemplates a computer readable medium containing instructions which cause the processor to perform any of the various techniques as described in this disclosure. In some cases, a computer readable medium may form part of a computer program product that can be sold to a manufacturer and/or used in a device. The computer program product can include computer readable media and, in some cases, packaging materials. The details of one or more aspects of the invention are set forth in the drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings. 135453.doc -12· 200934250 [Embodiment] FIG. 1 is a schematic diagram illustrating a video transmission system that supports video reconciliation. In the example of FIG. 1, the video transmission system 1G includes a source device 12 and a plurality of Destination sighs 14a, 14B (commonly "destination device 14"). $ " Also available for 12 to obtain digital video content from multiple sources, and to encode each video for transmission to destination device 14. The video content can, for example, be captured, archived (e.g., pre-fetched), computer generated, or combined. In each case, the video content can be encoded by the source device 12 and transmitted to the destination device via the communication channel. The source (4) u can include or be consuming to the 'transmitter, which includes appropriate radio frequency (RF) modulation, The filter and amplifier assembly delivers the encoded video via the communication channel by driving - or multiple antennas.

To support adjustable video, source device j 2 encodes the source video as a base layer stream (or base layer) and one or more adjustable enhancement layer bit streams (or enhancement layers). The base layer bitstream typically carries video data of a base quality level or multiple enhancement layers carrying additional video material referred to herein as refinement to support higher quality levels. The refinement coded in the enhancement layer can be incrementally increased, for example, by providing additional higher frequency coefficients or further refining existing coefficients.

Plus fidelity (for example, depending on the police product, L vengeance - in some examples, can be more reliable than the enhanced transmission (for example, with a lower packet error rate (PER)) as illustrated in Figure 1. The base layer and a single enhancement layer. One of the additional video data, in the source setting, for simplicity, the basis of the channel-channel, however, the source device 12 can encode the enhancement layer above the channel. In some examples 135453. Doc •13· 200934250 In-stream encoded source video to support different channels for selection by the user of the destination channel. Usually simultaneous transmission frequency

= Enable the destination device to select a different channel at any time for viewing with =. Because & ' is very similar to the TV viewing experience, the destination device _ user-selectable - channel to watch sports, and then select another - channel to watch the new m other scheduled broadcast program events. In general, 'each channel can be coded as—base layer and/or multiple enhancement layers. In addition, for illustrative purposes, quality adjustability (also known as signal-to-noise ratio (SNR) tunability) The technique of the present invention is described below. Then, the technology can be extended to spatial adjustability. In space-adjustability applications, the base layer carries video data at base-space resolution, and the enhancement layer carries additional video data to support higher spatial resolution. In some instances, system (7) may utilize video tunability that combines SNR, spatial and/or temporal adjustability. Source device 12 may encode source video as a base layer, e.g., according to the ITU-t h.264/MPEG 4 Part 1 〇 Avc standard SVC extension, and encode the source video as a reinforcement layer in accordance with the techniques described in this disclosure. Likewise, the techniques as described in this disclosure may be applied in some aspects to implement the video scalability extension of the device' which additionally conforms to the H.264 standard. In fact, the techniques of the present invention may represent potential modifications for future versions or extensions of the H.264 standard or other standards. However, these technologies can be used in conjunction with any of the other video compression standards, such as those defined in MPEG-1 and MPEG-2, ITU-T H.263 standard, film and television engineers. Association (SMPTE) 421M Video CODEC Standard (collectively referred to as "VC-1"), defined by the Chinese 135453.doc -14· 200934250 Audio Video Code Standard Working Group (collectively referred to as "avs. and by standard Any other video writing standard defined by the organization or developed by the organization as a proprietary standard. Destination device 14 may support wired and/or wireless reception of encoded video. Destination device 14 may include, for example, a wireless communication device capable of receiving and decoding. Any device for digital video data, such as cellular or satellite radiotelephone, wireless broadcast system, personal digital assistant (pDA), laptop computer, digital camera, digital recording device, video game device, video game Machine, digital television, digital live broadcast system, and the like, in the example of Figure 1, two destination devices 14A, 14B are shown. 10 may include any number of destination devices 14. Destination device 14 may also operate in accordance with any of the various video compression standards described above. Figure 1 illustrates destination device 14 relative to source device 12 that transmits encoded video. Positioning ^ In detail, the destination device 14A is closer to the transmission source (i.e., the source device 12 in Fig. 1), and the destination device 14B is far away from the transmission source. In the case of encoding the base layer with a lower PER, two The destination devices 4 8 and 146 can reliably receive and decode the base layer. The destination device 14A located closer to the source device 12 can also reliably receive the enhancement layer. However, the destination device 14B located away from the source device 12' It is possible, for example, that the enhancement layer cannot be reliably received due to network or channel conditions. Also 'because both the base layer data and the enhancement layer material are available, the closer destination device 14A can have higher quality video. And the destination device 14B can only present the minimum quality level provided by the base layer material. Therefore, the extra bits in the enhancement layer can be decoded and added to the base layer 135453.doc -15-200 The video obtained by the destination device 14 is adjustable in the sense of increasing the 1024250 bit stream to increase the signal-to-noise ratio (SNR) of the decoded video. However, the adjustability is only possible when the enhancement layer is present. Thus, the term "quality" as used in this invention may refer to objective and/or subjective visual quality. In other words, enhancement layer refinement results in video material being reproduced for higher quality of the original material. 'Enhanced video fidelity by enhancing layers. In other examples, network or channel conditions are sufficient for both destination devices 14A and 14B to receive the base layer and the enhancement layer. However, destination β devices 14 and (10) may have different decoder capabilities that prevent one of destination devices 14 and 14 from using the enhanced layer of overhead video data to produce higher quality video. If one of the destination devices 14 is a client device, such as a mobile handset or, for example, other small portable devices, there may be limitations due to computational complexity and memory requirements. Thus, destination device 14, which may have limited computation or memory resources, can design an adjustable video write code in a manner that only decodes the base layer. In this manner, destination device 14 with better network or channel conditions and/or higher decoder capabilities will be able to reconstruct video with higher video quality using additional video data from the enhancement layer. The techniques described in this disclosure utilize an entropy write technique that facilitates efficient code writing of enhancement layer bitstreams. The entropy coding technique of the present invention can be enabled to write, for example, additional video data in a refined form in a reinforced layer bit stream in a single coded one operation, thereby reducing write code complexity, write code delay, and memory. Body requirements. As will be described in further detail, source device 12 may, in some examples, encode the enhancement layer without knowing any subsequent coefficients (i.e., any coefficients of 135453.doc 200934250 that are currently being written by the non-zero coefficients). Each zero coefficient in the coefficient vector. Writing a code enhancement layer in a single operation eliminates the need for performing the first-time operation of the analysis coefficient vector and the second operation of writing the code coefficient vector based on the analysis. $coefficients For example - some conventional entropy encoders may perform a first-encoded-secondary operation 卩 to generate symbols for representing coefficient vectors, at least some of which represent more than one non-zero coefficient. In other words, it is necessary to know the subsequent coefficients to encode the non-zero coefficients of the vector. Additionally or alternatively, the f-entropy encoder may also select a VLC table for encoding symbols during the first or subsequent code encoding operation. In the -state, the ^^ table can be selected based on the generated symbols. Alternatively, the statistical data may be collected by analyzing the coefficient vector during the first encoding one operation, and the ¥1^(: table may be selected based on the collected statistical data. Then the second encoding is performed by a conventional entropy encoder One operation to entropy encode the coefficient vector based on the analysis performed during the first encoding one operation. As an example, some conventional entropy encoders may use selection based on generated symbols or other statistical data during the second encoding one operation. The VLC table encodes the symbols generated during the first operation. Generating symbols representing more than one non-zero coefficient and/or selecting a VLC table based on the generated symbols or other statistics may allow for more efficient coding of the coefficient vector. The technique of the present invention not only eliminates the need to use one or more encoding operations to compile the enhancement layer, but the entropy writing technique of the present invention can additionally result in writing code without storing and accessing coefficient information of the video material of the base layer. Enhance layers to further reduce computational complexity and memory requirements. Source device 12, destination device 14, or both A wireless or 135453.doc • 17-200934250 wired communication device as described above. Again, source device 12, destination device 14, or both may be implemented as an integrated circuit device (such as an integrated circuit die or wafer set), The integrated circuit device can be incorporated into a wireless or wired communication device or another type of device that supports digital video applications, such as a digital media player, a personal digital assistant (PDA), a digital television, or the like. A block diagram of source device 12 and destination device 14 of write code system 10 is described in further detail. Destination device 14 can be, for example, any of destination devices 14A or 14B of the figure. As shown in FIG. The source device 12 can include: a video source 18, a video encoder 2, and a transmitter 22. The video source 18 of the source device 12 can include a video capture device, such as a camera that has previously captured Videoconferencing of video or video feed from video content providers. As an alternative example, video source 8 can generate computer graphics based data as source video, or live video and computer generated video. In some cases, the source device 12 can be a so-called camera phone or video phone, in which case the video source 18 can be a camera. In each case, it is fetched, pre-fetched or computer generated. The video can be encoded by the video encoder 2 to be transmitted from the source device 12 to the destination device 14 via the transmitter 22 and the communication channel 16. The video encoder 20 receives video data from the video source 18 and encodes the video data into a base layer. The elementary stream and one or more enhancement layer bitstreams. In the example illustrated in Figure 2, video encoder 20 includes a base layer coder view-enhancement layer coder 32. base layer coder 3 〇 and reinforced The layer encoder, such as the video source 18, receives the shared video material. The base layer encoder encodes the video data at the first bit rate to generate the base layer of the video at the first quality level. 135453.doc 200934250 = two codecs 32 encoding Extra bits are used to generate - or multiple plus 1 force strong" " multiple enhancement layers provide the first when the video is enhanced to the second higher quality base layer when added to the base layer level, the reinforcement layer is Add to σ Xi stem, with the bit rate of 70, the second higher bit rate provides a higher, thus' may be considered as the reinforcing layer of the video coding, Ma refinement coding in the base layer. Refinement can, for example, be an additional factor and/or gradually increase the video of the existing system's refinement in the enhancement layer as it is decoded

The meaning of the quality of the data i 'the refinement of the coding in the enhancement layer can be hierarchical: thus the decoding of the refinement of all enhancement layers (for example) will result in the highest 70 rate and the highest mouth quality, but only The decoding of the refinement of the first enhancement layer will increase the bit rate and quality (4) relative to the decoding of only the base layer. The video data received from the video source 18 can be a series of video frames. The base layer encoder 30 and the enhancement layer encoder 32 divide the series of frames into write code units and process the write code units to encode the series of video frames. The codec unit can be, for example, an entire frame or a portion of a frame (such as a slice of a frame). The base layer encoder 30 and the enhancement layer encoder 3 2 divide each code writing unit into pixel blocks (referred to herein as video blocks or blocks) and operate on the video blocks in the individual code writing units so that Encode video data. Similarly, video material can include multiple frames. A frame can include multiple slices, and a slice can include multiple video blocks. The video blocks can be of fixed or varying size and can vary in size depending on the particular code standard. As an example 'ITU-T H.264/MPECM Part 10 AVC supports intra-frame prediction of various block sizes, such as 16 by 16, 8 by 8 or 4 by 4 for luminance components, and for chrominance components 8χ8, to 135453.doc -19- 200934250 and support inter-frame prediction of various block sizes, such as 16 times 16, 16 times 8, 8 times 16, 8 times 8, 8 times 4, 4 times 8 for the luminance component And 4 by 4 and the corresponding adjustable size for the chroma component. In H.264/MPEG-4 Part 10 AVC, each video block, commonly referred to as a macroblock (MB), can be subdivided into sub-blocks of fixed or varying size. That is, the code writing unit may contain sub-blocks having the same or different sizes. In general, MB and various sub-blocks can be considered as video blocks. Thus, the MB can be considered a video block, and the MB itself can be considered to define a video block set combination in the case of split or sub-segmentation. The encoders 30, 32 perform the in-frame writing and inter-frame writing of the video blocks of the frame. In-frame coding relies on spatial prediction to reduce or remove spatial redundancy in video material within a given code unit (e.g., 'frame or slice'). For in-frame coding, the encoders 30, 32 form spatial prediction blocks based on one or more previously encoded blocks within the same block as the block currently being coded. The prediction block can be a prediction pattern of the video block currently being coded. ❹ Base layer encoder 30 may perform interpolation (eg, according to the intra-frame write mode associated with the block) by using pixel values of one or more previously encoded blocks within the base layer of the current frame, for example. The prediction block is generated based on one or more previously encoded blocks within the frame. The enhancement layer encoder 32 may generate prediction blocks based on one or more previously encoded blocks within the frame. The enhancement layer encoder 32 may generate prediction blocks based, for example, on one or more previously encoded video blocks from the base layer and the enhancement layer within the frame. For example, enhancement layer encoder 32 may generate a prediction block using a weighted sum of pixel values from at least one previously encoded video block from the base layer and from the enhancement layer 135453.doc •20-200934250 At least one previously encoded video block. Daytime writes rely on time prediction to reduce or remove the time in the adjacent map of the video sequence. For inter-frame writing, the encoder performs motion estimation to track the movement of closely matched video blocks between two or more adjacent frames within the code unit. In the inter-frame prediction condition, the encoders 30, 32 may generate temporal prediction blocks based on one or more of the other coding blocks from within the code coding unit. The encoders 3, 32 may, for example, compare the current video block with the block of one or more adjacent video frames to identify the area of the adjacent frame that most closely matches the current video block. A block, for example, a block having a minimum mean square error (MSE), a squared difference sum (SSD), an absolute difference sum (SAD), or other difference measure in one or more adjacent frames. The encoders 30, 32 select the identified blocks in the adjacent frames as prediction blocks. The base layer encoder 30 compares the blocks of the current video block to the one or more adjacent frames of the base layer. The enhancement layer encoder 32 may compare the blocks of the current video block to one or more adjacent frames in the base layer and/or the enhancement layer. After the intra-frame or inter-frame prediction of the video block, the encoder 3, 32 generates the remaining block by subtracting the generated prediction block from the original video block being coded. The remaining block thus indicates the difference between the predicted block and the current block being coded. The encoders 30, 32 may apply transform, quantization, and entropy write procedures to further reduce the bit rate associated with communication of the remaining blocks. A transform technique that may include discrete cosine transform (DCT), integer transform, wavelet transform, direction transform, or other transform operation changes a set of pixel differences to residual transform coefficients, which represent pixel difference values in the frequency domain energy. The encoders 30, 32 apply quantization to the residual transform coefficients, which 135 453.doc - 21 · 200934250 often involves the process of limiting the number of bits associated with any given coefficient. The encoders 30, 32 scan the two-dimensional residual block to produce a one-dimensional vector of coefficients, and entropy encode the coefficient vector to further compress the remaining coefficients. Entropy coding may, for example, include variable length write code (VLC), arithmetic write code, fixed length write code, content adaptive VLC (CAVLC), content adaptive binary arithmetic write code (CABAC), and/or other entropy writes. Code technology. SNR adjustability can be achieved by residual quantization. In particular, base layer encoder 30 may quantize the remaining transform coefficients using a first quantization parameter (Qp), and enhancement layer encoder 32 may quantize the remaining transform coefficients using the second QP. In ITU-T H_264/MPEG-10 AVC, larger QPs typically result in the use of a smaller number of bits to encode video data at a lower quality, while smaller Qp results in a higher quality coded video with a larger number of bits. data. Thus, the base layer encoder encoding the video material with the minimum quality level can quantize the coefficients of the base layer using a Qp value greater than the Qp value used by the enhancement layer encoder 32 to quantize the coefficients of the enhancement layer. As a result, the quantized residual transform coefficients from the base layer encoder 3 represent the video sequence at the first quality, and the quantized residual transform coefficients from the enhancement layer encoder represent additional coefficients or fine coefficients of the existing coefficients of the video sequence. The additional coefficients or refinements increase the quality of the video sequence to a second, higher quality when combined with the base layer. The 'codes II 30' 32 each receive a -dimensional coefficient vector representing the quantized residual transform coefficients of the base layer and the enhancement layer, respectively. In other words, the base layer encoder 30 receives the coefficient vector of the base layer, and the enhancement layer coder 32 receives the coefficient vector of the corresponding enhancement layer. Although the encoders 30, 32 receive the same original video material ', the coefficient vectors can be different. This may be due to the base layer encoder 30 and the enhancement layer coder 32 that produce different 135453.doc -22-200934250 prediction blocks, for example, the base layer encoder 30 from one or more previously compiled base layers The block generates a prediction block, and the enhancement layer encoder 32 generates a prediction block from - or a plurality of previously encoded base layer blocks and enhancement layer blocks. The base layer encoder 30 and the enhancement layer coder 32 each encode respective coefficient coefficients i to respectively generate a base layer bit stream and at least a enhancement layer bit stream. The base layer encoder 3Q and the enhancement layer encoder 32 encode different coefficient vectors using different write code techniques according to the technique of the present month. The base layer encoder 3 may encode a coefficient vector using a plurality of (four) encoding-secondary operations, in which the base layer encoder 30 analyzes the coefficient vector during at least one of the marsh-time operations, And based on the analysis, the coefficient vector is encoded during at least one subsequent encoding operation. In an example, the base layer encoder 3 may encode the quantized residual transform coefficients of the base layer coefficient vector according to cavlc as defined in the H.264/MPEG-4 Part 10 AVC standard. The cAvLc as defined in the H.264/MPEG-4 Part 10 AVC standard can encode the base layer coefficient vector using a plurality of coding one operations. During a first coded one operation, base layer encoder 30 may generate symbols to represent coefficient vectors, at least some of which represent more than one non-zero coefficient, and in some cases, all coefficients of the coefficient vector. Base layer encoder 30 may, for example, be based on (10) symbiotic symbols as defined in the H 264/MpEG 4 Part 1 AVC Standard, which represent the total number of coefficients in the coefficient vector ("Totalc〇effs"), The number of trailing ] in the coefficient vector ("Tls"), the sign of any trailing 1 sign, the magnitude (or level) of the non-zero coefficient except the trailing 1 , and the sum of all the moves ("sumRuns" And each of the 135453.doc -23· 200934250 non-zero coefficients. In order to generate some of the symbols, such as TotalCoeff and sumRuns, the base layer encoder 30 can analyze the entire coefficient vector. During the first encoding operation, The base layer encoder 30 may also select a VLC table based on analysis of the coefficient vector for use during subsequent encoding operations. In some examples, the base layer encoder 30 may generate based on one operation during the first write code. The symbol selects the VLC table for use during a subsequent (eg, second) encoding operation. For example, the base layer encoder 3 selects the VLC table for the total number of coefficients (TotalCoeffs) in the block. Used when encoding the sumRuns symbol, because there is a relationship between these two values. In detail 'As TotalCoeffs increases, sumRuns decreases, and as TotalCoeffs decreases, sumRuns increases. Again, based on the block The total number of coefficients (TotalCoeffs) selects the VLC table to use the allowable base layer encoder 30 to select the VLC table that more efficiently encodes the sumRuns when encoding the sumRuns symbol. It can be used for other symbols to be encoded or using other collected statistics. Performing a similar VLC table selection. The base layer encoder 30 encodes a symbol (T〇talC〇eff) representing the total number of non-zero coefficients in the coefficient vector and a number indicating the trailing 1 during the second or other subsequent encoding operation. Symbol (called Tls). The number of trailing 1 is the number of coefficients having a magnitude of 1, and when the coefficient vector is read in reverse order (ie, starting from the end of the coefficient vector), the coefficients have A coefficient greater than 1 occurs before the occurrence of a coefficient vector. The base layer encoder 3 选择 can select a VLC table based on the predicted number of non-zero coefficients to encode TotalCo Use the 'eff and T1 symbols' and use the selected vLC table to compose the 135453.doc -24-200934250 code TotalCoeff and T1 symbols. Select the VLC table based on the predicted number of non-zero coefficients to use in the coded TotalCoeff and Τ1 symbols. The base layer encoder 30 is allowed to select a VLC table that more efficiently encodes the TotalCoeff and T1 symbols. Thus 'different VLC tables can be more efficient for different predicted numbers of non-zero coefficients. In an example, the base layer encoder 3 may predict the current block based on the number of non-zero coefficients of one or more previously encoded video blocks (eg, upper adjacent video blocks and left adjacent video blocks) The number of non-zero coefficients. 〇 The base layer encoder can encode any trailing 1 sign. For example, for each of the trailing 1s, the base layer encoder 3〇 can encode "1" (if the trailing 1 sign is positive), and encode "〇" (if trailing The sign of 1 is negative. Thus, the base layer encoder 30 may not need to perform VL C table selection for the sign. The base layer encoder 3 〇 can encode the magnitude of the non-zero coefficient other than the trailing one. The layer encoder 3 may encode a level of non-zero coefficients using a vlc table, a fixed length write code, or other type of write code. For example, the base layer encoder 30 may encode a non-zero 'coefficient using a binary write code. The base layer encoder 30 may encode symbols (sumRuns) representing the number of zero-valued coefficients occurring in the coefficient vector before the last non-zero coefficient. As described above, the base layer encoder 30 may be based on the block. The total number of coefficients (TotalCoeffs) selects the VLC table to use when encoding the sumRuns symbol because there is a relationship between these two values. The base layer encoder 30 can encode the last non-zero coefficient from the coefficient vector starting at each Before a non-zero coefficient The lifetime (or length of the run). The length of the run length 135453.doc •25· 200934250 is the number of zero-valued coefficients before the non-zero coefficient. Therefore, the base layer encoder 30 can first encode the fortune before the last non-zero coefficient of the coefficient vector. The length (i.e., the number of 'zero value coefficients'), followed by the length of the run length before the previous non-zero coefficient, etc., up to the length of the run length before the first non-zero coefficient of the coding coefficient vector. The base layer encoder 30 can select the VLC table. Used to separately encode each of the lengths of the journey. The base layer encoder 3〇 can be based on the sum of the tolls

The (SFmRuns) symbol and the sum of the coded hits to date select the VLc table to encode the current run value. As an example, if the coefficient vector has a runsum of (SumRuns) and the run length encoded before the last non-zero coefficient of the code is six, then all residual runs must be zero, one or two. Since the length of the possible path is gradually shortened as each additional process is encoded, the base layer encoder 30 can select a more efficient VLC table to reduce the number of bits used to represent the fortune. In this way, the base layer encoder 3 performs an operation to encode the base layer coefficients 'the one-time operation code includes an analysis base layer remaining block, the coefficient vector (for example) to generate symbols and/or select vlc The first-time operation of the table and the second code-based operation based on analyzing the coding coefficient vector. Although the base layer encoder 30 is described above as being used as in H.264/MPEG_, points! (10) defined in the Q AVC standard "codes the quantized residual transform coefficients, but the base layer encoder 30 may encode the quantized residual transform coefficients using other write coding methods. The enhancement layer encoder 32 encodes the formable coefficients. The quantized residual transform of the enhancement layer of the vector form, the enhancement layer coding (4), may produce a quantized residual coefficient of the quantized residual coefficients of the 135453.doc -26 - 200934250 different from the base layer. The quantized residual coefficient of the enhancement layer is attributed to The different QPs may be used during quantization and may be different from the quantized residual coefficients of the base layer. In addition, the quantized residual transform coefficients may be different from the quantized residual transform coefficients of the base layer, since the remaining blocks represent the original video blocks and The difference between the prediction blocks generated by the previous coding blocks forming the base layer and the enhancement layer. The remaining blocks of the base layer are the original video block and the prediction block generated using the previously coded block only from the base layer. The difference between the layers can thus include additional coefficients and/or & refinement of existing coefficients. In this sense, the video in the enhancement layer The quantized residual transform coefficients of the block represent the refinement of the video blocks encoded with the first quality in the base layer and provide higher quality video material when added to the base layer.

The enhancement layer encoder 32 may discard one or more of the quantized residual coefficients of the coefficient vector during encoding, depending on the available bit rate. For example, when zigzag scanning is used to complete the coefficient scan as illustrated in Figure 3, enhancement layer coder 32 may discard coefficients corresponding to the high frequency transform basis function, e.g., bits _ near the end of the coefficient vector. The quantized residual coefficients according to the CAVLC encoding as defined in the H-Part 10 AVC standard may not allow the enhancement layer encoder 32 to discard the coefficients, & because the symbols to be encoded are in the 'some' (for example, T〇talC 〇effs& sumRuns) refers to all the coefficients in the block. If the enhancement layer encoder 32 discards - or more of the coefficients of the coefficient vector, the received information will be redundant, thus resulting in lower code efficiency. In addition, because the decoder must receive the progression of all non-zero coefficients in the block to be able to properly solve 135453 in zigzag scanning when encoding using CAVLC as defined in the H 264/MpEG 4 Part 1 AVC Standard. Doc -27· 200934250 The position of each coefficient, so the enhancement layer encoder 32 may not be able to discard the coefficients of the coefficient vector from the enhancement layer. Thus, enhancement layer encoder 32 encodes the coefficients of the enhancement layer in accordance with the write code technique of the present invention. The enhancement layer encoder 32 encodes the quantized residual transform coefficients of the coefficient vector in a single coded one operation. In other words, enhancement layer encoder 32 does not perform the first-time operation to analyze the coefficient vector, and does not subsequently encode the symbol during the second operation based on the analysis. The fact is that the enhancement layer encoder 32 starts from the beginning of the coefficient vector and encodes each of the non-zero coefficients one by one in a single encoding operation. In this way, the enhancement layer encoder 32 can analyze the coefficient vector without In the case of any subsequent coefficients (i.e., without knowing any subsequent coefficients of the coefficient vector) each of the non-zero coefficients is encoded.

In one aspect, enhancement layer encoder 32 may encode, for each of the non-zero coefficients, a symbol indicating that there is at least one residual non-zero coefficient in the coefficient vector. The symbol may, for example, be the end of the block ( Ε0Β) symbol. The enhancement layer encoder 32 can encode symbols using a single bit. For example, enhancement layer tarcoder 32 may encode zeros in the presence of at least one residual non-zero coefficient (eg, soil > although non-zero coefficients), and may encode each in the absence of residual non-zero coefficients After the ΕΟΒ symbol of the coefficient, the enhancement layer encoder 32 encodes the run before the current non-zero coefficient. As described above, the motion represents the number of zero-valued coefficients occurring between the beginning of the coefficient vector (1) of the previous non-zero coefficient or coefficient vector (in the case of the first non-zero coefficient) and the current non-zero coefficient. The enhancement layer encoder 32 can encode the run using a single vlc table. In one of your % examples, at 135453.doc -28- 200934250

When TotalCoeffs is equal to one, the enhancement layer encoder 32 can encode the run to write the code sumRuns using the VLC table used in the CAVLC as defined in the H-264/MPEG-4 Part 10 AVC standard. In other words, the enhancement layer encoder 32 can reuse one of the vLC tables that have been maintained by the video encoder 2〇. In other examples, enhancement layer encoder 32 may encode the motion using one of the other VLC tables that have been maintained by video encoder 20. Alternatively, enhancement layer encoder 32 may maintain a separate VLC table that is specifically designed to encode the coefficient vector of the enhancement layer. In either case, enhancement layer encoder 32 ® may not need to adaptively select a VLC table for use in Coding the journey. The fact is that the enhancement layer encoder 32 can use a single VLC table, thus eliminating the need for a first-order operation to collect statistics for selecting VLC tables. The enhancement layer encoder 32 encodes the sign of the non-zero coefficient after the encoded run of each coefficient. The enhancement layer encoder 32 may, for example, encode "丨" (if the sign of the non-zero coefficient is positive), and encode "〇" (if the sign of the non-zero coefficient is negative). The enhancement layer encoder 32 may borrow The value of the non-zero coefficient is set to a team towel to adjust the magnitude of the non-zero coefficient. In some examples, the enhancement layer coding (4) may not encode the magnitude of the non-zero coefficient. In this way, the enhancement layer is programmed. The value of the non-zero coefficient can be limited to one. The destination device 14 is then configured to decode all of the non-zero coefficients identified in the refinement to have a magnitude equal to one that does not encode the coefficients of the enhancement layer. This can result in some loss of peak-to-noise ratio (pSNR), but reduces the number of bits used to encode the coefficients. In this manner, enhancement layer encoder 32 can, for example, be unaware of any subsequent coefficients in the coefficient vector. The coefficients of the enhancement layer bitstream are encoded in a single operation. Since the enhancement layer encoder 32 does not need to analyze the coefficients to 135453, doc-29-200934250 quantities, for example, to produce one or more non-zero coefficients representing the vector. Symbol or choice V The LC table encodes symbols, so only one encoding operation is performed. Conventional encoders typically perform at least two one-time operations; (1) analyze the first operation of the coefficient vector' and (2) the second one based on analyzing the coding coefficient vector In addition, the enhancement layer encoder 32 may encode the coefficients of the enhancement layer using a single VLC table' thus eliminating the need to perform an encoding one operation for forming symbols for adaptively selecting the codebook table. The enhancement layer encoder 32 can reduce write code complexity, write code delay, and memory requirements. In addition, the write code technique of the present invention can additionally result in code enhancement without storing and accessing coefficient information of the base layer. The coefficients of the layers further reduce computational complexity and memory requirements. The source device 12 transmits the encoded video material to the destination device 14 via the transmitter 22. The destination device 14 can include the receiver 24, the video decoder 26, and the display. Apparatus 28. Receiver 24 receives the encoded video bitstream from source (4) via channel 16. As described above, the encoded video bitstream includes a base layer A bit stream and/or a plurality of enhancement layer bit streams. The video decoder % decodes the base layer and (if available) one or more enhancement layers to obtain video data. In particular, the video decoder 26 includes - base layer decoding 34 and an enhancement layer decoder 36. The base layer decoder 34 decodes the base layer bitstream received via channel 16 to produce video material of the first quality for presentation on display device 28. Enhancement layer deblocker Decoding a bitstream of one or more enhancement layers to obtain additional video material (eg, refinement) that increases the quality of the decoded video material to a second, higher quality. The number of enhancement layers received by device 14 (e.g., -, two, '135453.doc -30. 200934250 or more) depends on visual channel conditions or other restrictions. Moreover, the number of enhancement layers received by the enhancement layer decoder 36 can be determined by the decoder limitations. The encoding and decoding of the base layer combined with the selected number of enhancement layers permits the SNR quality of the decoded video. Incremental improvements.

The base layer decoder 34 decodes the base layer to obtain a symbol representing the vector of the quantized residual coefficients of the base layer. The base layer decoder 34 can decode the base layer to obtain the total number of non-zero coefficients in the block, the number of trailing 1s of the block, the trailing 1 plus and minus sign, the magnitude of the coefficient other than the trailing 1, all the fortune The fortune before the sum of β and the non-zero coefficient. The base layer decoder W may further decode the base layer bitstream to identify the VLC table used to decode the base layer symbols. In other examples, base layer decoder 34 may select a VLC table for use based on previously decoded symbols. Using the decoded symbols, the base layer decoder 34 can reconstruct the coefficient vectors of the base layer. The enhancement layer decoder 36 decodes the bit stream of the enhancement layer to obtain, for example, a vector of extra residual coefficients or an enhancement of the enhancement layer of the refined form of the existing residual coefficients. In particular, enhancement layer decoder 36 decodes the duration and sign of the enhancement layer coefficients using the same VLC table as the VLC table used by enhancement layer encoder 32 until the EOB symbol indicates that no non-zero coefficients are remaining. Using the decoded symbols, enhancement layer decoder 36 reconstructs the coefficient vectors of the enhancement layer blocks. The decoders 34, 36 reconstruct each of the blocks of the coded unit using the decoded quantized residual coefficients. After generating the coefficient vector, the decoders 34, 36 inversely scan the coefficient vector to produce a two-dimensional block of quantized residual coefficients. The decoders 34, 36 inverse quantize (ie, dequantize) the quantized residual coefficients 135453.doc 200934250 and apply an inverse transform to the dequantized residual coefficients (eg, inverse DCT, inverse integer transform, inverse wavelet transform, or inverse direction transform) ) to generate the remaining blocks of pixel values. The decoders 34, 36 sum the prediction blocks generated by the decimators 34, 36 and the remaining blocks of pixel values to form reconstructed base layer video blocks and enhancement layer video blocks, respectively. The base layer video block and the enhanced layer video block are combined to form a video block having a higher resolution. The decoders 34, 36 generate © prediction blocks in the same manner as described above with respect to the encoders 30,32. The destination device 14 can display the reconstructed video block to the user via the display device 28. Display device 28 can include any of a variety of display devices, such as a cathode ray tube (CRT), liquid crystal display (LCD), plasma display, light emitting diode (LED) display, organic LED display, or another type of display. unit. In some examples, video encoder 20 and video decoder 26 are configured to provide an adjustable boost bit stream that can be arbitrarily truncated. Therefore, the system can avoid the use of discrete enhancement layers that must be coded as a whole to achieve adjustability. However, in some embodiments, system 10 can be configured to support adjustability using, for example, a generalized fine grain size adjustability (FGS) method or a discrete enhancement layer on a selective basis. Source device 12 and destination device 14 can operate in a substantially symmetrical manner. For example, source device 12 and destination device 14 may each include a video encoding and decoding component. Thus, system 1 can support, for example, one-way or two-way video transmission between devices 12, 14 for video streaming, video broadcasting, or video telephony. 135453.doc -32- 200934250 In some aspects 'for video broadcasting, the techniques described in this disclosure can be applied to enhanced H.264 video writing for use in forward link only ( FLO) Air Intermediary Specification, the forward link empty intermediary specification for terrestrial mobile multimedia multicast, (as the technical standard tia_i〇99 ("fl〇 specification") published in July 2007) Instant video services are delivered in the Mobile Multimedia Multicast (tm3) system. That is, communication channel 16 may include wireless information channels for broadcasting wireless video information in accordance with the FLO specification or the like. The FLO specification includes an example of demarcating the meta-stream syntax and semantics and a decoding process suitable for several short 中介 intermediaries. Alternatively, the video can be broadcast according to other standards such as DVB_H (Palm Digital Video Broadcasting), ISDB-T (Ground Integrated Services Digital Broadcasting) or DMB (Digital Media Broadcasting). Thus, source device 12 can be a mobile wireless terminal, a video streaming server, or a video broadcast server. However, the techniques described in this disclosure are not limited to any particular type of broadcast, multicast or peer-to-peer system. In the broadcast condition, the source device 12 can broadcast a number of frequency channels of the video material to a plurality of destination devices, each of which can be similar to the destination device 14 of FIG. Thus, although a single destination device 14 is shown in Figure 1, for video broadcasts, source device 12 will typically broadcast video content to many destination devices simultaneously. In other examples, the wheeler 22, the communication channel 16 and the receiver 24 can be configured to communicate in accordance with any wired or wireless communication system, including one or more of the following: Ethernet, Telephone (eg, POTS) H power line and fiber optic system and or wireless system, the wireless system comprising one or more of the following: a coded multi-directional proximity (CDMA or 135453.doc • 33-200934250 CDMA2000) communication system , frequency division multi-directional proximity (FDMA) system, orthogonal frequency division multi-directional (OFDM) proximity system, time-sharing multi-directional proximity (TDMA) system (such as GSM (Global System for Mobile Communications), GPRS (General Packet Radio Service) Or EDGE (Enhanced Data GSM Environment)), TETRA (Terrestrial Relay Radio) Mobile Phone System, Wideband Coded Multi-Directional Near-Connect (WCDMA) System, High Data Rate lxEV-DO (First Generation Evolutionary Data) or lxEV -DO Gold Multicast System, IEEE 402.18 System, MediaFLOTM System 'DMB System, DVB-Η System, or another solution for data communication between two or more devices. Video encoder 20 and video decoder 26 may each be implemented as one or more microprocessors, digital signal processors (DSPs), special application integrated circuits (ASICs), field programmable gate arrays (FPGAs), discrete logic, Soft palate, hard body, firmware or any combination thereof. Each of video coder 20 and video masher 26 may be included in one or more encoders or decoders, either of which may be in a respective mobile device, user equipment, broadcast device, server, or The similarity is integrated into the part of the combined encoder/decoder (C〇DEC). In addition to this, source device 12 and destination device 14 each may include appropriate modulation, demodulation, frequency conversion, filtering, and amplification components for transmitting and receiving encoded video, as applicable, including sufficient to support wireless communication. Radio frequency (RF) wireless components and antennas. However, for ease of illustration, such components are summarized as transmitter 22 of source device 12 and receiver 24 of destination device 14 in FIG. 3 is a block diagram illustrating an example base layer encoder 3 and enhancement layer encoder 32 in further detail. In the example of FIG. 3, the base layer encoder includes a prediction unit 33A, a frame storage 35A, a transformation unit 38A, a quantization unit 135453.doc • 34-200934250, a coefficient scanning unit 41A, and an inverse quantization unit 42. Inverse transform unit 44A, base layer entropy encoder 46 and summer 48a and "sumer 48". The different features in Figure 3 are intended to highlight different functional aspects of the illustrated device, and do not necessarily imply that such elements must be implemented by separate hardware or software components. The truth is that the functionality of one or more units can be integrated into a separate hardware or software component. • Predicted single TO 33A uses in-frame or inter-frame prediction to generate prediction blocks. The prediction block can be a prediction pattern of the current video block being coded. As described above, prediction unit 3 generates prediction blocks from intra-prediction using one or more previously encoded blocks of the base layer within the same frame as the block currently being coded. Alternatively, the prediction unit may generate the prediction block using inter-frame prediction based on one or more of the previous coding blocks in one or more of the base layers. Prediction unit 33A may extract the previous coded block from frame storage 35. After the intra-frame or inter-frame prediction of the video block, the base layer encoding artist 30 generates the remaining area by subtracting the prediction block generated by the prediction unit 33A from the current video block at the summer 48A. Piece. The remaining blocks include a set of pixel difference values that quantify the difference between the pixel values of the current video block and the pixel values of the predicted block. The remaining blocks can be represented in a two-dimensional block format (e.g., a two-dimensional matrix or array of pixel values). In other words, the remaining blocks are two-dimensional representations of pixel values. Transform unit 38A applies a transform to the remaining blocks to produce a residual transform coefficient. Transform unit 38A can, for example, apply DCT, integer transform, direction transform, wavelet transform, or a combination thereof. After applying the transform 135453.doc -35· 200934250 to the remaining blocks of the pixel values, the quantization unit 40A quantizes the transform coefficients to further reduce the bit rate. After quantization, inverse quantization unit 42A and inverse transform unit 44A may apply inverse quantization and inverse transform, respectively, to reconstruct the remaining blocks. The summer 48B combines the reconstructed residual block with the predicted block generated by the prediction unit 33 A to generate a reconstructed video block for storage in the frame memory 358. The reconstructed video blocks stored in the frame store 34 can be used by the prediction unit 32 of the base layer encoder 30 to write subsequent video blocks in-frame or inter-frame. In addition, as will be described in more detail below, the reconstructed video block stored in the frame store 35A can be used by the prediction unit 33B of the enhancement layer encoder 32 to block the video blocks in the in-frame or inter-frame code enhancement layer. Refinement. After the quantization, the coefficient scanning unit 41A scans the coefficients from the two-dimensional block format into a one-dimensional vector format. This process is generally referred to as coefficient scanning. The coefficient scanning unit 41A can scan the two-dimensional block of coefficients, for example, using a zigzag scanning order as described in further detail in FIG. After scanning, the base layer entropy encoder 46 entropy encodes the coefficients of the one-dimensional vector. The base layer encoder "can entropy encode coefficient coefficients, for example, using CAVLC as defined in the H.264/MPEG-4 Part 10 AVC standard and described above in detail with respect to Figure 2. The enhancement layer encoder 32 includes a prediction. Unit 33B, frame storage 35B, transform unit 38B, quantization unit 4〇B, coefficient scanning unit *丨b, inverse quantization unit 42B, inverse transform unit 44B, a boost layer entropy encoder 49, and summers 48C and 48D ("summer 48"). The elements of the enhancement layer encoder 32 are substantially similar to the similarly numbered units of the base layer encoder. Thus, only the differences will be described. The prediction unit 338 of the enhancement layer encoder 32 generates a prediction block of the 135453.doc • 36 · 200934250 prediction pattern of the current video block. Unlike the prediction unit 33 A of the base layer encoder 30 that uses only the previous coding block of the base layer to generate the prediction block, the prediction unit 33B of the enhancement layer encoder 32 may be based on one or more previous coding of the base layer and the enhancement layer. The block produces a prediction block. In other words, prediction unit 33B can generate predicted blocks using reconstructed video blocks from the base layer and reconstructed video blocks from the enhancement layer. For example, prediction unit 3 3 b may combine the reconstructed blocks of the reconstructed video block and the enhancement layer of the base layer to produce a second higher quality prediction block. β because the prediction block generated by the prediction unit 33B is based on the base layer and

The enhancement layer is generated by reconstructing the video block, so the remaining blocks generated at the summer 48C represent the current video block and the self-base layer and the enhancement layer construction (ie, in the second higher visual quality). The difference between the previously coded blocks. Although operatively similar to the quantization unit 4A of the base layer encoder 3, the quantization unit 4B of the enhancement layer encoder 32 can use different Qp to quantize the transform coefficients. As described above with respect to Figure 2, adjustability can be achieved by using different quantization parameters. For example, when the base layer encoder 3 and the boost layer encoder 32 operate in accordance with ITU_T H.264/MpEGiG, the measure π40Α can encode the video beaker using a sand value greater than the value used by the quantizing unit. The result 'the quantized 2-coordinate transform coefficients from the base layer encoder 表示 represent the video sequence in the first quality, and the residual transform coefficients from the booster bed i H quantize represent the extra coefficients or the refinement of the existing coefficients of the view. These additional coefficients or refinements increase the product f of the video sequence to a second higher visual quality when compared to the base «. I35453.doc -37- 200934250 Further, as described in detail with respect to FIG. 2, the enhancement layer entropy encoder 49 encodes the quantized residual transform coefficients in a single code-sub-plane. In other words, the enhanced layer entropy processor 49 can encode each non-zero coefficient of the coefficient vector of the enhancement layer without knowing any subsequent Wei of the coefficient vector. Writing a code enhancement layer in a single operation eliminates the need to perform a first-of-analysis coefficient vector--: and the second-time operation for writing code coefficient vectors based on analysis. The reality is that the enhanced layer encoder 49 starts at the beginning of the coefficient vector and encodes each of the coefficients one by one in a single encoding operation. More details regarding the entropy coding of the enhancement layer are described in relation to Figure 4. 4 is a block diagram illustrating an example base layer entropy encoder 46 and a boost layer entropy encoder 49 in further detail. The base layer entropy encoder 46 may include an analysis single 7L 50, a plurality of VLC tables 52A to 52N ("VLC^ 52"), a total coefficient encoder core trailing I (Tls) encoder 56, a sign The encoder 58, a coefficient magnitude encoder 6, and a motion encoder 62, and a run length encoder 64. The enhancement layer entropy encoder 49 may include an E〇B symbol encoder 66, a run length encoder 68, a positive and negative encoder 7〇, and a table 69. The base layer entropy encoder 46 encodes the coefficient vector representing the video block of the first quality by performing a plurality of encoding one operations. According to a CAVLC as defined in the H.264/MPEG-4 Part 1 AVC standard, for example, the base layer entropy encoder 46 may perform an analysis coefficient vector to, for example, generate a symbol representing a coefficient vector and/or select a VLC table. The first encoding one operation and the second encoding one operation based on analyzing the encoding coefficient vector. As an example, the analysis unit 5 of the base layer network encoder 46 can analyze the number 135453.doc -38 - 200934250 number vector to produce one or more symbols representing the coefficient block. The analyzing unit 5 can determine, for example, the number of total coefficients (TotalCoeff) in the block, the number of trailing i (T1), and each trailing 1 according to the H.264/MPEG-4 Part 1 AVC standard. Positive and negative signs, the magnitude of each non-zero coefficient, the sum of the operations (SumRuns), and the length of the run before each non-zero coefficient. At least some of the symbols produced by analysis unit 50 (e.g., T〇taic〇eff and sumRims) may represent all coefficients of the coefficient vector. The analysis unit 5 can generate more symbols or fewer symbols in other examples. Additionally or alternatively, analysis unit 50 may select a subset of VLC tables 52 for encoding symbols during the first or subsequent encoding operations. In one aspect, analysis unit 50 can select a subset of vlC table 52 based on the generated symbols. Alternatively, analysis unit 50 may collect statistical data during analysis of the coefficient vectors and select a subset of VLC tables 52 based on the collected statistics. For example, base layer encoder 30 may select a VLC table based on the total number of coefficients in the block (TotalCoeffs) to use when encoding the sumRuns symbol, since _ is a relationship between the two values. As will be described in greater detail below, selecting a subset of VLC tables 52 based on the generated symbols or other statistical data may enable more efficient encoding of the symbols representing the coefficient vectors. The base layer masher 46 encodes the coefficient vector during a second or other subsequent write code operation. In particular, the total coefficient encoder 54 encodes the total number of non-zero coefficients (TotalCoeff) in the coefficient vector. The total coefficient encoder 54 may encode TotalCoeff using the one selected in the VLC table 52 based on the prediction of the number of non-zero coefficients of the current coefficient vector. In one example, the current number of non-zero coefficients may be based on a number of non-zero coefficients that bite a plurality of previously encoded video blocks (eg, an upper adjacent video block and a left 135453.doc -39 - 200934250 side neighboring video block) Prediction of the number of non-zero coefficients of the coefficient vector. In this way, the base layer entropy decoder can select the same VLC table based on previously decoded blocks. After the total coefficient encoder 54 encodes the total number of non-zero coefficients, the τ 1 s coder 56 encodes the Tls symbol. The Tls encoder 56 may encode the Tls symbol, for example, using one of the VLC tables 52 based on the predicted number of non-zero coefficients, in the same manner as described above with respect to the total coefficient encoder 54. The sign encoder 58 encodes the sign of any trailing 1s. For example, © s, for each of the trailing ones, the sign encoder 58 can encode 1" (if the trailing 1 sign is positive) and encode (if the trailing sign is negative). The system of value encoders 60 encodes the level of the trailing non-zero coefficients (eg, magnitude). The coefficient magnitude encoder 6 can be encoded using a VLc table, a fixed length code, or other type of entropy code. The level of the non-zero coefficient. The run and encoder 62 may encode a symbol representing the number of zero-valued coefficients occurring in the coefficient ❹ vector before the last non-zero coefficient, i.e., 3111111111113 symbols. The run and encoder 62 uses a block-based block. The sum of the coefficients in the total (TotalCoeffs) selects one of the VLC tables 52 to encode the sumRuns character. In addition, the VLC table is selected based on the total number of coefficients in the block (T〇talc〇effs) to encode the sumRuns symbol. Use the Allow Run and Encoder to "select a VLC table that encodes sumRuns more efficiently. The run length encoder 64 encodes the run length of the coefficient vector. The long-running encoder 64 may first encode the run length of the last non-zero coefficient of the coefficient vector, followed by the run length of the previous-non-zero coefficient, etc., until the first non-zero coefficient of the coding coefficient to 135453.doc -40-200934250 Until the length of the previous journey. In other words, the longitude encoder can begin to compile the final run length first. The run length encoder 64 may encode each of the run lengths using a VLC table 52 selected based on the sum of the total motions of the coefficient vectors (嶋 嶋 )) and the sum of the coded runs to date. As an example, if the coefficient vector has a run length of eight and (SUmRUnS) and the run length coded before the last non-zero coefficient of the code is six, the residual mave must have zero, one or two. Since the length of the possible run length is gradually shortened as the #帛 outer pass is encoded, the long run encoder 64 can select a more efficient VLC table to reduce the number of turns used to represent the run: In this manner, the VLC table 52 used by the run length encoder is variable for each of the run lengths. The enhancement layer entropy encoder 49 in the single-code-time operation represents, for example, an additional coefficient or a coefficient vector of the video block for the refined form of the existing coefficients to form a reinforcement layer. As will be described in further detail, the source device 12 may, in some examples, encode each non-zero coefficient in the coefficient vector of the strong layer without knowing any subsequent coefficients. The enhancement layer coder starts at the beginning of the coefficient vector and is encoded in the single coding operation. In this way, the enhancement layer encoder 49 encodes the coefficient vector on a per-coefficient basis without the coefficients occurring in the coefficient vector. Write the code in the single-time (four) plus (4) the first-time operation of the (four) = ΓΓ number vector and the second operation for writing the code coefficient vector based on the analysis. For each of the 2 zero coefficients, there is at least a residual non-zero coefficient in the EGB (four) coded paper coding day non-coefficient vector 135453.doc • 41 - 200934250 for example and 3, EOB symbol encoder 66 Zero may be encoded in the presence of at least one residual non-zero coefficient (eg, at least the current non-zero coefficient), and one may be encoded when there are no more residual non-zero coefficients. After the _ symbol of each coefficient is compiled, the run length encoder 68 encodes the run length before the non-zero coefficient. As described above, the run length represents the number of zero-valued coefficients of the current non-zero coefficient. The run length encoder uses a single VLC table 69 to encode the run length. In one example, the vlc table 69 can be the same as one of the base layer entropy encoders 46 of the table 52. Or 〇 I 'the run length encoder 68 can maintain a separate VLC table that is specifically designed to encode the coefficients of the enhancement layer's coefficient vector. In either case, the run length encoder 68 may not need to adaptively select the VLC table for the coded run. Rather, the run length encoder 68 can use a single VLC table, thus eliminating the need for a first operation to collect statistics for selecting a VLC table. After the encoded run length of each coefficient, the sign encoder 7〇 encodes the sign of the non-zero coefficient. The sign encoder 7〇 can, for example, encode 1 (if the sign of the non-zero coefficient is positive) and encode “〇" (if the sign of the non-zero coefficient is negative). The enhancement layer entropy encoder 49 may not encode stronger. The magnitude of the non-zero coefficient of the layer, which can result in a loss in peak signal-to-noise ratio (pSNR), but reduces the number of bits used to encode the coefficients. The trick-writing code technique of the present invention allows for enhanced layer entropy coding The coefficients of the enhancement layer bitstream are encoded in a single-time operation. Since the enhancement layer entropy encoder 49 does not analyze the coefficient vector (for example) to generate symbols and/or select a vLc table, only one encoding operation is required. Conventional encoders usually perform at least two 135453.doc -42- 200934250

= sub-operation: (1) the first operation of the analysis coefficient vector, and (7) the second operation of the base two-resolution coding coefficient vector... Bu, the enhancement layer smoke--49 can use the single-VLC table to encode the coefficients of the enhancement layer, thus eliminating The lamp is used to make the selection of the various VLC tables for the purpose of the operation. Strengthening the layer entropy encoder 49 by 'methods' can reduce write code complexity, write code delay, and memory requirements. Further, the entropy write code technique of the present invention can additionally result in the case where coefficient information of the base layer is not stored and accessed. Write down the coefficients of the code enhancement layer to further reduce computational complexity and memory requirements. Figure 5 is a block diagram illustrating an example of base layer decoder 34 and enhancement layer decoder 36 in further detail. The base layer decoder "includes a base layer entropy decoder 72, a coefficient scanning unit 74", an inverse quantization unit 76, an inverse transform unit 78, a prediction unit 80, a frame memory 82, and a summer 84. The enhancement layer decoder 3.4 includes a enhancement layer entropy decoder 86, a coefficient scanning unit 74A, an inverse quantization unit 76A, an inverse transform unit 78A, a prediction unit, a frame storage 82A, and a summer 84. The base layer entropy decoder 72 decodes the received base. Layer bitstream to generate video material of a first quality for presentation on a display device. Base layer entropy decoder 72 receives the base layer bitstream and decodes the base layer bitstream to obtain (eg, quantized residual coefficients) Remaining information in one dimensional vector form and header information (eg, in the form of one or more header syntax elements). Base layer entropy decoder 72 performs base layer entropy encoder 46 by FIGS. 3 and 4. The reciprocal decoding function of the encoded code is performed. In detail, the base layer entropy decoder 72 decodes the base layer to obtain a symbol representing the vector of the quantized residual coefficients of the base layer. In use as in 135453.doc • 43 · 200934250 When the CAVLC defined in the H, 264/MPEG-4 part of the i〇AVC standard is coded, for example, the base layer entropy decoder 72 can decode the base layer to obtain the total number of non-zero coefficients (TotalCoeff) in the block, The number of trailing 1s (Tls) of the block, the trailing 1 sign, the magnitude of the coefficient other than the trailing 1 , the sum of all the runs (sumRuns), and the foreman before the non-zero coefficient. In some examples, the VLC table selected for decoding may be selected based on previously decoded blocks or previous decoded symbols of the current block. In other examples, 'base layer entropy decoder 34 may decode the base layer bit stream for identification. The VLC table of the base layer symbols is decoded. Using the decoded symbols, the base layer decoder 34 may reconstruct the coefficient vectors of the base layer. After generating the coefficient vectors, the coefficient scanning unit 74A inversely scans the coefficient vectors to produce quantized residual coefficients. The two-dimensional block. The inverse quantization unit 76A inverse quantizes (ie, 'de-quantizes) the quantized residual coefficients, and the inverse transform unit 78A applies an inverse transform to the dequantized residual coefficients (eg, inverse dct, inverse Transform, inverse wavelet transform, or inverse direction transform) to generate residual blocks of pixel values. Prediction unit 80A uses one or more adjacent blocks within a common frame or predicted between frames in the case of intra-frame prediction. The prediction block is generated by using one or more blocks in the adjacent frame. The prediction unit generates the prediction block using only the previous coding block from the base layer. The summer 84A is generated by the prediction unit 80A. The prediction block is summed with the remaining blocks of pixel values to form a reconstructed base layer video block. The base layer video block is stored in the frame store 82A for use in generating subsequent prediction blocks. The enhancement layer decoder The bit stream of the enhancement layer is decoded to obtain, for example, a vector of the outer residual coefficients of the amount I35453.doc -44· 200934250 or a refinement of the video data to the refined form of the existing residual coefficients. The enhancement layer entropy decoder 86 decodes the duration and sign of the enhancement layer coefficients using the same VLC table as the VLC table used by the enhancement layer entropy encoder 49 until the E〇B symbol indicates that the non-zero coefficients are no longer residual. The symbol, enhancement layer entropy decoder 86 reconstructs the coefficient vector of the enhancement layer block. The decoded coefficient vector represents additional bits that do not increase the quality of the decoded video material to the second higher quality refinement when combined with the bits of the base layer. After the coefficient vector is generated, the coefficient scanning unit 74B reverse scan coefficient

The vector is used to produce a two-dimensional block of quantized residual coefficients. Inverse quantization unit 76B inverse quantizes (i.e., dequantizes) the quantized residual coefficients, and inverse transform unit 78B applies an inverse transform to the dequantized residual coefficients (e.g., inverse dct, inverse integer transform, inverse wavelet transform, or inverse direction transform). ) to generate the remaining blocks of pixel values. Prediction unit 80B uses one or more neighboring blocks within a common frame or one or more blocks within an adjacent frame to generate a predicted block in the case of intra-frame prediction or in the case of inter-frame prediction. . The prediction unit generates prediction blocks using previously encoded blocks from both the base layer and the enhancement layer. Summer 84B sums the prediction blocks generated by prediction unit 80B with the remaining blocks of pixel values to form a reconstructed enhancement layer video block. The enhancement layer video block is stored in the frame store 82B for use by the prediction unit 80B to generate subsequent prediction blocks. The reconstructed base layer video block and the reconstructed enhancement layer video block are combined at summer 84C to form a higher quality video block. 135453.doc • 45· 200934250 FIG. 6 is a block diagram illustrating the example base layer decoder 72 and the enhancement layer entropy decoder 86 in further detail. The base layer entropy decimator 72 may include a plurality of VLC tables 52A to 52N ("VLC table 52"), a total coefficient decoder 9A, a trailing l(Tls) decoder 92, and a sign decoder 94. A coefficient magnitude decoder 96, a and a range decoder 98, and a run length decoder 丨〇〇. The enhancement layer entropy decoder 86 may include an EOB symbol decoder 102, a run length decoder 104, and a sign. Decoder 106 and a Vlc table 69. The base layer entropy decoder 72 decodes the base layer bitstream to obtain symbols representing the coefficient vectors of the video blocks at the base πσ quality level. The total coefficient decoder 90 uses the VLC table 52. One of them decomposes the bitstream to obtain the total number of non-zero coefficients in the coefficient vector (TotalCoeff). The total coefficient decoder 9〇 may be based on a prediction of the number of non-zero coefficients of the current coefficient vector (eg, based on one or more The VLC table 52 for decoding TotalCoeff is selected for the number of non-zero coefficients of the previously decoded video blocks. In this way, the total coefficient decoder 9A can be selected to be the same as the VLC table used by the total coefficient encoder 54. VLC table 52 to encode TotalCoeff After the total coefficient decoder 90 decodes the total number of non-zero coefficients, the Tls decoder 92 decodes the Tls symbol. The Tls symbol represents the number of coefficients having a magnitude of one, and when the coefficient vector is read in reverse order, The coefficients are encountered before a coefficient greater than a magnitude. The T1 s decoder 92 can decode the Tis symbol using one of the VLC tables 52 selected based on the predicted number of non-zero coefficients. The sign decoder 94 decodes Any trailing sign of 1. For example, the sign decoder 94 can determine that the sign of the coefficient is positive for each of the trailing ones when receiving 135453.doc -46 - 200934250 to "1" And when the "0" is received, the sign of the coefficient can be determined to be negative. The coefficient magnitude decoder 96 decodes the magnitude of the non-zero coefficient except the trailing one. The coefficient magnitude decoder 96 can use the VLC table, A fixed length write code or other type of entropy write code to decode the level of the non-zero coefficient. The run and decoder 98 may decode the symbol representing the number of zero value coefficients occurring in the coefficient. vector before the last non-zero coefficient, ie sumRuns• The transport and decoder 98 decodes the sumRuns symbol © using one of the VLC tables 52 selected based on the total number of coefficients in the block (T〇talCoeffs), and the total number of coefficients (TotalCoeff) in the block is determined by the total coefficient. The decoder 9 performs the previous decoding. Again, the VLC table is selected based on the total number of coefficients in the block (T〇talc〇effs) to use the allowable run and decoder 98 to select a more efficient decoding table when decoding the sumRuns symbol. The length decoder 100 decodes the length of the run of the coefficient vector. The run length decoder 100 may first decode the run length of the last non-zero coefficient of the coefficient vector, followed by the run length of the previous non-zero coefficient, etc., until the run length before the first-non-zero coefficient of the decoded coefficient φ eigen. In other words, the run length decoder 100 can first begin decoding the last run length. The long run length "Decoding 1164 can be based on the M_Runs of the total distance of the coefficients to 4" and the animal table 52 selected by the sum of the coded runs so far, and each of the lengths of the run length is decoded. The forward and decoder 98 previously decoded the _^ character. However, the 'transport length decoder 1' can collect statistical data on the sum of the so-called decoded to date. Since the length of the possible run length becomes shorter as each additional run is decoded, the run length decoder (10) can select a more efficient VLC table to reduce the number of bits used to represent the run. 135453.doc • 47_ 200934250 The VLC table 52 used by the run length decoder 100 in this manner is variable for each of the operational lengths. The enhancement layer entropy decoder 86 decodes the bitstream of the enhancement layer to obtain, for example, a vector of additional coefficients or refinement of the video blocks in a refined form of the existing coefficients. The EOB symbol decoder 1〇2 determines whether the EOB symbol indicates whether there is at least one residual non-zero coefficient. When there is at least one residual non-zero coefficient, the run length decoder 104 decodes the run length before the next non-zero coefficient. The run length decoder 104 can decode the run length of the next non-zero coefficient using the same VLC table 69 as the VLC table used by the run length encoder 68. The sign encoder 106 decodes the sign of the non-zero coefficient. For example, the sign encoder 106 can determine that the sign of the coefficient is positive (when "1" is received') and negative (when "" is received). The enhancement layer entropy decoder 86 continues to decode the non-zero coefficients until the E0B symbol decoder 1〇2 indicates that there are no residual non-zero coefficients. FIG. 7 is a conceptual diagram illustrating a zigzag scan of a 4x4 coefficient block 40. The zigzag scan shown in Figure 7 can be implemented by the encoders 3, 32 of Figure 2. The scan order of the zigzag scan shown in Figure 7 follows the arrows through the video block 11 and is labeled in the scan order by a factor (1 to (; 16). In detail, it is shown in Figure 7. The value indicates the position of the coefficient within the sequential one-dimensional vector and does not represent the actual value of the coefficient. The result of the zigzag scan shown in Figure 7 is the one-dimensional coefficient vector JT, where [cl, c2, c3, C4, c5 , c6, c7, c8, c9, clO, ell, cl2, cl3s cl4, cl5, 〇16] where cl to cl 6 represent coefficient positions within a two-dimensional array of coefficients. 135453.doc • 48- 200934250 It is not limited to any particular scanning order or technique. For example, the scanning order used in the present invention may be the zigzag scanning order shown in Fig. 7. Alternatively, the scanning order used in the present invention may be other scanning order, such as 'Horizontal scanning, vertical scanning or any other scanning technique. Figure 8 is a conceptual diagram illustrating a hypothetical example of a coefficient block 12 of the coefficients of the enhancement layer. In this example, the values shown in Figure 8 indicate the coefficients at the locations. Actual value of the coefficient block 120 The coefficient values may represent quantized residual coefficients, unquantized transform coefficients, or other types of coefficients of the video block in the enhancement layer. In the example illustrated in Figure 8, the coefficient block 丨2〇 is a 4x4 block. The techniques of the present invention can be extended to apply to blocks of any size. After the coefficient block 120 is scanned according to the zigzag scan illustrated in Figure 3, the resulting coefficient vector κ is: Factory = [4, 0, 0 , -2, 〇, 1. The enhancement layer encoder 32 encodes each of the coefficients of the coefficient vector K in accordance with the techniques described in this disclosure. As an example, for each of the non-zero coefficients of the coefficient vector factory For example, the enhancement layer encoder 32 encodes the E〇B symbol, the run length, and the sign. As described in detail above, the E〇B symbol indicates whether there are any residual non-zero coefficients in the coefficient vector, and the run length is represented at the current of the coefficient vector. The number of zero-valued coefficients that occur before the non-zero coefficient, and the positive or negative sign indicates whether the coefficient value is positive or negative. According to the aspect of the present invention, the enhancement layer encoder 32 may not encode the magnitude of the coefficient. Floor Encoder 32 may encode each of the non-zero coefficients as if all non-zero coefficients were equal to 1. In this way 135453.doc • 49· 200934250, the enhancement layer encoder 3 2 can be considered as encoding the following The coefficient vector 〆, and V 〇 wide = [1, 〇, 〇, -1, 0, 1, 〇, 〇, 〇, 〇, 0, 0, 0, 0, 〇, 〇] ❹ The enhancement layer encoder 32 can (for example) using the code group equal to zero, the code group of zero, and the positive and negative ones equal to one to compile the first system (ie, 4 in the coefficient vector F or the coefficient vector, 丨) Encoding the second coefficient (ie, _2 in the coefficient vector F or _ in the coefficient vector wide) using e〇b equal to zero, a codeword group of two fortune, and a sign equal to zero, and equal to zero Then, the codeword group of the fortune and the sign of equal to - followed by the delete symbol equal to - encode the third non-zero system, the number (ie, the coefficient vector 〆 or the coefficient vector ~ U ° as above The set of VLC tables defined in the H.264/MPEG.4 Part 10 AVC standard is obtained for encoding the group of the fortune. An example encoded bit stream is described for illustrative purposes. Reinforcement layer coding = can be coded in different ways without departing from the material of the invention. For example, the _ symbol can be encoded as an additional (four) coefficient in the block of Table 2, and encoded as zero to indicate no residual non-zero. Similarly, the sign can be coded to zero to represent a positive non-zero coefficient, and encoded as _ to represent a negative non-zero system, number. As an alternative, the _ symbol for each-non-zero coefficient encoding may be the most unique of the vector. You may not have an EOB symbol at the encoded bit stream. (4) Non-zero pure n. The sinus, the helmet, the production number indicates the current system, and when the moxibustion is non-zero coefficient, the spectroscopy code H knows that there is no extra coefficient of the block after decoding the current coefficient of 135453.doc -50· 200934250. A flowchart illustrating an example operation of a video encoder (such as the video coder 20 of FIG. 2) that implements the adjustable video coding technique of the present invention. The base layer encoder 30 and the enhancement layer of the video encoder 20 The encoder 3 2 obtains the video material (130) from the video source 18. As described above, the base layer encoder 3 and the enhancement layer encoder 32 obtain the same original video data. The video data obtained from the video source 8 can be ( For example) is a series of video frames.

For each video block, the base layer encoder 3 encodes the base layer (132) using a write code technique that performs multiple coded one operations. The base layer encodes the video block with the first quality level. The base layer encoder 3 产生 can generate a coefficient vector representing the video block of the first quality and encode the remaining transform coefficients of the block to generate the base layer. The base layer encoder 3 may generate a base layer according to a cavlc coding coefficient vector as defined in the H.264/MPEG-4 part i 〇 AVC standard. As described in detail above with respect to Figure 2, base layer encoder 30 may perform a first __code-time operation to analyze the coefficient vector and a second operation to analyze the coefficient vector based on the analysis. For each video block, enhancement layer encoder 32 encodes the extra bits into enhancement layer (134) using a write code technique that performs a single code-second operation. Enhancing the extra bit coding refinement of the layer bit stream, which enhances the video to a second higher quality level when added to the base layer bit stream, although in this example the enhancement layer encoder 32 is described as Only (four) single-layer 'but plus (four) encoding (four) can encode more than one enhancement horizon. In this case, the reinforcement layer can be hierarchical in the sense that the enhancement layer provides a progressively higher:= as it is decoded. INTEGRATION 135453.doc 51 200934250 The second entropy writing technique used by the enhancement layer encoder 32 encodes the e〇b symbol, the operation and the positive for each of the non-zero coefficients _ of the coefficient vector of the enhancement layer. As described in detail above, the EOB symbol can indicate whether any residual non-zero coefficients are present, the run length represents the number of zero-valued coefficients that occurred before the non-zero coefficients, and the sign indicates whether the coefficient value is positive or negative. After the last non-zero coefficient is positively shaped, the (4) code (4) encodes the e〇b symbol to indicate that there are no residual non-zero coefficients. ❹

The base layer coded H3G and the enhancement layer encoder 32 separate the coded base layer bit stream and the enhancement layer bit stream (i 36). The drop-on-mother technique used by the enhancement layer encoder 32 allows the residual coefficients of the enhancement layer to be encoded with less computation and implementation complexity without significant write efficiency. The entropy coding technique of the present invention can enhance the write code of the additional video data in the layered bit stream in a single coded operation, for example, in a single coded one operation, thereby reducing write code complexity, write code delay and Remember (4) requirements. For example, the enhancement layer encoder 32 may encode the coefficients of the enhancement layer to each non-zero coefficient without knowing any (four) coefficients, thereby allowing the vector of the coefficient to be written in the single-second operation order, and eliminating the execution - for analyzing the coefficients The first of the vectors - the second operation and the second - order operation based on the analysis of the code system 'number vector. Figure 10 is a flow diagram showing an example operation of a enhancement layer encoder (such as the enhancement layer encoder of Figure 2) of the residual coefficients of the video enhancement block of the coding enhancement layer in accordance with the present invention. The enhancement layer coder identifies the first non-zero coefficient (140) in the enhancement layer region j _ vector. The enhancement layer encoder 32 encodes the coefficient vector of the enhancement layer block to have at least one residual non-zero system B symbol (142). The enhancement layer encoder 32 may encode a 135453.doc -52 - 200934250 EOB symbol using a single bit, e.g., 'code zero when there is at least a money non-four number and code one when there are no residual non-zero coefficients. The enhancement layer encoder 32 encodes a run (144) indicating the number of zero value coefficients preceding the non-zero coefficients. The enhancement layer encoder 32 may encode the flight in a number of examples using a stored VLC table as defined in the H.264/MPEG-4 Part 1 AVC standard. For example, the enhancement layer encoder uses a VLC table to encode the run when the total number of coefficients (T〇talc〇effs) is equal to the sum of the total runs of the code (sumRuns). Alternatively, enhancement layer encoder 32 may maintain a separate VLC table that is specifically designed to encode the coefficient vectors of the enhancement layers. The enhancement layer encoder 32 may encode the sign of the non-zero coefficient (146). The enhancement layer encoder 32 may, for example, encode "if the sign of the non-zero coefficient is positive", and encode "0" (if the sign of the non-zero coefficient is negative). In some examples, the enhancement layer encoder 3 2 may not encode the magnitude of the non-zero coefficient. In this manner, the enhancement layer encoder 32 may limit the magnitude of the non-zero coefficient to one. Thus, any non-zero coefficient having a magnitude greater than one is set equal to 1. The magnitude of the non-zero coefficient of the non-encoded enhancement layer may result in a certain loss of the peak-to-noise ratio (PSNr) but reduces the number of bits used to encode the non-zero coefficient. The enhancement layer encoder 32 determines the enhancement layer. Whether there are any residual non-zero coefficients in the block (148). When there is at least one residual non-zero coefficient in the enhancement layer block, the enhancement layer encoder 32 continues to encode the EOB, fortune of each of the residual non-zero coefficients And a sign. When there are no residual non-zero coefficients in the enhancement layer block, the enhancement layer encoder 32 encodes the EOB symbol to indicate that there are no residual non-zero coefficients in the coefficient vector of the enhancement layer block (149). 135453.doc -53- 2 00934250 describes that the enhancement layer is transmitted along with the base layer. Because the enhancement layer write code technique described in Figure ί does not write code; and the symbol of more than one coefficient, the enhancement layer write code technique allows for the available bit rate. The enhancement layer encoder 32 discards - or more of the quantized residual coefficients of the coefficient vector during encoding. In addition, the enhancement layer write coding technique reduces write code complexity and implementation. Figure 11 illustrates the decoding enhancement layer bit stream. To obtain the direction of the remaining transform coefficients

Flowchart of an example operation of a quantity of enhancement layer decoders, such as enhancement layer decoder 36 of FIG. The enhancement layer decoder 36 obtains a enhancement layer bitstream (150). The enhancement layer decoder 36 analyzes the EOB symbols to determine if there are any residual non-zero systems 1 5 2 ). The enhancement layer decoder 36 may, for example, determine the presence of a ☆-residual non-zero coefficient when the decimation symbol is equal to zero, and determine that there are no residual non-zero coefficients when the ε 〇 β symbol is equal to one. When enhancement layer decoder 36 determines that there is at least a residual non-zero coefficient (e.g., the 'Ε〇Β symbol is equal to zero), enhancement layer decoder 36 decodes the motion associated with the next non-zero coefficient (154). The number of zero-valued coefficients associated with the lower-non-zero coefficient, the non-zero system, and the number W. The enhancement layer decoder uses a table-like (four) νιχ table that is encoded by the enhancement layer (4) to decode the run. In the example - 'enhancement layer decoding (4) can be used for CAVL (as in the H.264/MPEG-4 part 1) 〇 丄 I knife AVC standard (for the total code for the code and (sumR deleted) The VLC table (in the system, the number (4) (T. (four). effs) is equal to -) is used to decode the run. However, other tables can be used as long as it is the same table used by the enhancement layer encoder 32. The enhancement layer decoder 36 sets the number of coefficients of the run length equal to the non-zero coefficient to equal 135453.doc • 54· 200934250 at zero (156). If the run length is equal to two, then for example, the enhancement layer decoder 36 The two coefficients before the non-zero coefficient can be set equal to zero.

The enhancement layer decoder 36 decodes the sign of the non-zero coefficient (丨58). The sign of a non-zero coefficient can be decoded as positive (when the sign is equal to one) and decoded as negative (when the sign is equal to zero). After decoding the sign of the non-zero coefficient, the enhancement layer decoder 36 may set the non-zero coefficient equal to positive one or negative one based on the decoded sign (丨60). As described above, the enhancement layer may not encode the enhancement layer. The magnitude of the coefficient. Likewise, enhancement layer decoder 36 can be configured to set the magnitude of all non-zero coefficients equal to one. The enhancement layer decoder 36 continues to decode the run and sign of the non-zero coefficients until the enhancement layer decoder 36 determines that there are no residual non-zero coefficients (e.g., the E〇B symbol is equal to one). At this point, if any coefficients are left, the enhancement layer decoder 36 sets the residual coefficient s of the vector equal to zero. As described in detail with respect to Figure 2, the enhancement layer decoder % uses coefficient vectors and other data in addition to the prediction blocks to reconstruct the video blocks for presentation to display 28. 12 through 15 are block diagrams illustrating different configurations of encoders and/or decoders in an adjustable video write. These example encoders and decoders are illustrative of encoders that may utilize the techniques of the present invention. Types of. However, the actual configuration should in no way limit the techniques as described. Technology can be used in any adjustable video encoder.

Illustrated in FIG. 12 to FIG. 15 each of the video encoders and decoders can utilize the entropy code writing technique described in the present invention to promote the enhancement layer bits & Effective code writing. The entropy write 1 technique of the present invention may be difficult to perform in the single-code-under-operation for the additional form of the refined layer in the enhanced layer bit stream (for example) in the refined form 135453.doc • 55- 200934250 Write code 'to reduce write code complexity, write code delay and memory requirements as detailed in the step-by-step description of any coefficient after the non-zero coefficient of the code being written without knowing any subsequent coefficient (ie *) In the case of coding the per-non-zero coefficient in the coefficient vector of the enhancement layer. Writing a code enhancement layer in a single-time operation eliminates the need to perform a first-time operation of an analysis coefficient vector and a second-time operation for writing a code coefficient vector based on analysis. FIG. 12 is a block diagram illustrating an example adjustable video encoder 170. The ❹ 胄 video encoder 17Q can, for example, correspond to the video encoder 2G of FIG. In the example of FIG. 12, the 'adjustable video encoder 17' includes a base layer coding benefit 30'. The base layer encoder 3 includes a prediction unit 172, a frame storage unit 173, a transform unit 174, and a quantization unit. And β75β, inverse quantized το 176A and 176B, inverse transform unit 177, multiplexer module 178, and summers 179A to 179C. (d) The different features described in 3 are intended to highlight the different functional aspects of the device described, and do not necessarily imply that such elements must be implemented by separate hardware or software components. In fact, the functionality associated with one or more units can be integrated into a common or separate hardware or software component. The prediction list A 172 generates prediction blocks using in-frame prediction or inter-frame prediction. The prediction block can be a prediction pattern of the current video block being coded. As described above, the prediction unit 172 may generate the prediction block using intra-frame prediction based on one or more & pre-coded blocks of the base layer within the same frame as the block currently being coded. Alternatively, the prediction unit may generate an 135453.doc - 56 - 200934250 prediction block using inter-frame prediction based on one or more previously coded blocks within one or more adjacent frames of the base layer. Prediction unit 172 can extract the previously encoded block from frame storage 173. After the intra-frame or inter-frame prediction of the video block, the base layer encoder 30 generates the remaining block by subtracting the prediction block generated by the prediction unit 172 from the current video block at the summer 179A. . The remaining blocks include a set of pixel difference values that quantify the difference between the pixel values of the current video block and the pixel values of the predicted block. The remaining blocks can be represented in a two-dimensional block format (e.g., a two-dimensional matrix or array of pixel values). In other words, the remaining block is a two-dimensional representation of the pixel 〇 value. Transform unit 174 applies a transform to the remaining blocks to produce a residual transform coefficient. Transform unit 174 can, for example, apply DCT, integer transform, direction transform, wavelet transform, or a combination thereof. After applying the transform to the remaining blocks of pixel values, the quantization unit 17 5 A quantizes the transform coefficients to further reduce the bit rate. The output of quantization unit 175 A corresponding to the quantized coefficients associated with the base layer is provided to multiplex module 178. _ After quantization, inverse quantization unit 176A applies inverse quantization to produce a reconstructed version of the remaining blocks of the transform coefficients. The original residual block of the transform coefficients output by the summer transform unit 1 74 is subtracted from the reconstructed version of the remaining block of the transform coefficients output from the inverse quantization unit 1 76 A. This block, referred to herein as a transform difference block, is provided to quantization unit 175B. Quantization unit 1 75B quantizes the transform coefficients to further reduce the bit rate. An output of the quantized unit 丨75β corresponding to the quantized coefficients associated with the enhancement layer is provided to the multiplex module 178. In an example, quantization unit 175A may quantize the residual coefficients using the first QP, and quantization unit 175B may quantize the residual system 135453.doc -57 - 200934250 by the second qP. The second QP can be, for example, half the value of the first Qp, i.e., QP/2. After quantization by quantization unit 175B, inverse quantization unit 176B applies inverse quantization to produce a reconstructed version of the transformed difference block. The summer 179C sums the reconstructed version of the remaining block of the transform coefficients output by the self-inverse quantization unit 176 and the reconstructed version of the transformed transform block that is output by the inverse quantization unit 丨7 to produce a reconstructed residual Block. Inverse transform unit 177 applies an inverse transform to the reconstructed version of the video block. The beta reconstructed video block is stored in the frame store 1 < 73 and can be used by the predictive list 7L 172 to write the subsequent video blocks in-frame or inter-frame. Prediction unit 172 can provide control data to multiplex module 178, such as motion vector, uncut size, in-frame write mode, or the like. Multi-tasking module 丄 Combine base layer data and reinforcement layer data. In some examples, multiplex block 178 may include an entropy encoder for entropy encoding base layer data and enhancement layer data. In other examples, the base layer encoder and the enhancement layer encoder can be independent of the multiplex module.

I Figure 13 is a block diagram illustrating an example adjustable video decoder 18A. The adjustable video decoder 180 can, for example, correspond to the video decoder 26 of FIG. The adjustable video decoder 18 of FIG. 13 includes a demultiplexing module "I, inverse quantization units 182A and 182B, an inverse transform unit 183, a prediction unit 184, a frame memory 185, and a summer 186 eight and eighteen. The demultiplexing module 1 81 receives the adjustable encoded video and demultiplexes the multiplexed signal. In some examples, the demultiplexing module 181 may include an entropy decoder for entropy decoding the base layer data and the enhancement layer data. In other examples, the base layer solution 135453.doc -58·200934250 code and enhancement layer decoder can be inversely quantized (ie, dequantized) associated with the base layer independently of the demultiplexing module ‘ The residual coefficients are quantized, and inverse quantization unit 182B dequantizes the quantized residual coefficients associated with the enhancement layer. In an example, inverse quantization unit 182A may quantize the residual coefficients using the first QP, and inverse quantization unit 丨 82B may use A QP quantizes the residual coefficient difference. The second qp can be, for example, half the value of the first Qp, that is, QP/2. The dequantized transform output by the inverse quantization units 182A and 182B at the summer 186 Λ Individual sets of coefficients are added The reconstructed residual transform block. As described above, the dequantized transform coefficients output by the inverse quantization unit 182A may correspond to the basic quality level, and the dequantized transform coefficients output by the inverse quantization unit 182B are added. The output to the inverse quantization unit 182B results in an increased quality level. The inverse transform early element 183 applies an inverse transform to the sum of the dequantized residual coefficient blocks (eg, inverse DCT, inverse integer transform, inverse wavelet transform, or inverse direction transform). To generate the remaining blocks of pixel values. The summer 丨 86B adds the prediction block generated by the prediction unit 184 to the remaining blocks of the pixel values to form a reconstructed base layer video block. As described in detail above, The predicting unit 184 may generate the prediction region using one or more adjacent blocks in the common frame or using one or more blocks in the adjacent frame in the case of intra-frame prediction or in the case of inter-frame prediction. The block may be stored in the frame store 185. Figure 14 is a block diagram illustrating another example of an adjustable video decoder 19. The adjustable video decoder 190 may, for example, be viewed from Figure 2 The decoder 26 corresponds. The adjustable video decoder 19 of FIG. 14 includes a demultiplexing module 191, inverse quantization units 192A and 192B, inverse transform units 193A and 193B, and a prediction list 135453.doc -59-200934250 yuan 194, The frame storage 195 and the summer 196 are 198. The demultiplexing module 191 receives the adjustable encoded video and demultiplexes the signal. In some examples, the 'demultiplexing module 1 8' may include entropy. Entropy decoder for decoding base layer data and enhancement layer data. In other examples, the base layer decoder and the enhancement layer decoder may be independent of the demultiplexing module. Inverse quantization unit 192A and inverse transform unit 193A apply inverse quantization (i.e., dequantization) and inverse transform operations to the decoded residual coefficients associated with the base layer to obtain a reconstructed version of the remaining blocks of the base layer. Inverse quantization unit 〇 192B and inverse transform unit 193B apply inverse quantization (i.e., dequantization) and inverse transform operations to the decoded residual coefficients associated with the enhancement layer to obtain a reconstructed version of the remaining blocks of the enhancement layer. In an example, inverse quantization unit 192A may quantize the residual coefficients using the first qp, and inverse quantization unit 192 may quantize the residual coefficient differences using the second QP. The second Qp can be, for example, half the value of the first, i.e., QP/2. Prediction unit 194 may generate prediction regions using one or more blocks within the adjacent frame using one or more neighboring blocks within the common frame or in the case of inter-frame prediction in the case of intra-frame prediction. The prediction block can be stored in the frame storage 195. The summer 196 相 combines the prediction block-block generated by the prediction unit 194 with the reconstructed residual block output from the inverse transform unit 193 产 to produce decoded video material at a basic quality level. The self-adjustable video decoder 190 outputs decoded video material having a basic quality level. Decoded video material having a basic quality level is also provided to summer 196B. The summer 196B combines the output of the summer 196A with the reconstructed version of the remaining block of the enhancement layer output from the inverse transform unit 193B to produce a 135453.doc 200934250 for the second higher quality level of the solution to the solution. data. The self-adjustable video decoder 190 outputs decoded video material having a basic quality level. Figure 15 is a block diagram showing another example video encoder 2A. In the example of FIG. 15, the base layer encoder 3A includes a prediction unit 33A, a frame storage 35A, a transformation unit 38A, a quantization unit 4A, a coefficient scanning unit 41A, an inverse quantization unit 2A, and an inverse transformation unit. Pepper, base layer entropy encoder 46, summers 48A through 48C, and in-frame prediction unit 4A. Portraying the different features in Figure 3 as a unit intended to highlight the different functional aspects of the illustrated device, Φ ^• does not necessarily imply that such elements must be implemented by separate hardware or software components. The fact is that the functionality associated with one or more units can be integrated into a common or separate hardware or software component. Prediction unit 33A uses the inter-frame prediction (e.g., motion compensated prediction) to generate a prediction block. The prediction block can be the prediction pattern of the current video block being coded. As described above, prediction unit 33A may generate prediction blocks based on one or more previously encoded blocks in one or more adjacent frames of the base layer using inter-frame prediction techniques. Prediction unit 33 may extract the previously encoded block from frame storage 35A. After the inter-frame based prediction of the video block, the base layer encoder 3 generates the remaining block by subtracting the prediction block generated by the prediction unit 33a from the current video block at summer 48A. The remaining block includes a set of pixel difference values that quantify the difference between the pixel value of the current video block and the pixel value of the predicted block. The remaining blocks may be represented in a two-dimensional block format (e.g., a two-dimensional matrix or array of pixel values). In other words, the remaining blocks are two-dimensional representations of pixel values. 135453.doc 61 200934250 Transform unit 3 8 A applies a transform to the remaining blocks to produce the residual transform coefficients. Transform unit 38A can, for example, apply DCT, integer transform, direction transform, wavelet transform, or a combination thereof. After applying the transform to the remaining blocks of pixel values, quantization unit 40A quantizes the transform coefficients to further reduce the bit rate. After quantization, inverse quantization unit 42A and inverse transform unit 44A may apply inverse quantization and inverse transform, respectively, to reconstruct the remaining blocks. The summer 48B adds the reconstructed residual block to the predicted block generated by the prediction unit 33A to produce a reconstructed video block for storage in the frame storage 35A. The reconstructed video block stored in the frame memory 34 can be used by the prediction unit 32 of the base layer encoder 3 to write the subsequent video blocks in-frame or inter-frame. In addition, 'the reconstructed video block stored in the frame memory 35A can be used by the prediction unit 33B of the enhancement layer encoder 32 to block the video blocks in the in-frame or inter-frame code enhancement layer as will be described in more detail below. Refinement. After quantization, summer 48C subtracts the intra-frame predicted blocks generated by in-frame prediction unit 40A from the quantized residual coefficients. The in-frame prediction unit 40A may generate a prediction block using intra-frame prediction based on a block of the same coding block as the block currently being coded. The base layer entropy encoder 46, for example, uses the coefficient p enhancement layer coding output from the summer 48C as defined in the H.264/MPEG-4 Part 1 AVC Standard and described above with respect to Figure 2 for CAVLC. The processor 32 includes a prediction unit 33B, a frame storage 35B, a transform unit 38B, a quantization unit 40B, a coefficient scanning unit 41B, an inverse quantization unit 42B, an inverse transform unit 44B, a boost layer entropy encoder 49, and summers 48D to 48F. . The elements of the enhancement layer encoder 32 are substantially similar to the similarly numbered units of the base layer encoder 135453.doc-62-200934250. Thus, only the differences will be described. The prediction unit 33B of the enhancement layer encoder 32 generates a prediction block that is a prediction pattern of the current video block. Unlike the prediction unit 33 A of the base layer encoder 3 that uses the previous coding block of the base layer to generate the prediction block, the prediction unit 33B of the enhancement layer encoder 32 may be based on one or more previously coded blocks of the enhancement layer The prediction block is generated. The reconstructed video block of the enhancement layer may be at a second quality level above the quality level of the prediction block of the base layer. An additional difference between the enhancement layer encoder 32 and the base layer encoder 30 is that the output of the inverse quantization unit 42B of the enhancement layer encoder 32 and the inverse quantization unit 42A of the enhancement layer encoder 30 are at the summer 48F. Output combination. Adding the outputs of the inverse quantization units 42A and 42B produces a higher quality reconstructed video block, thus allowing for better prediction by the prediction unit described above. The techniques described in this disclosure can be implemented in hardware, software, firmware, or any combination thereof. Any feature described as a unit or component can be implemented together in an integrated logic device or separately as a discrete but interoperable logic device. If implemented in software, the techniques can be at least partially implemented by a computer readable medium containing instructions. When the instructions are executed, the methods described above are performed. The computer readable medium can form part of a computer program product' which product can include packaging materials. Computer readable (4) may include random access (RAM) such as synchronous dynamic random access memory (SDRAM), read only memory (R〇M), non-volatile random (suggest, electrically erasable) Money _eprqm), fast = memory, magnetic or optical data storage media, and the like. In addition (4) 135453.doc -63- 200934250 computer-readable communication of the form code of the order or data structure, such technology can be Media that carries or delivers a finger and that can be accessed, read, and/or executed by a computer is at least partially implemented.

The code may be executed by - or multiple processors, such as - or multiple digital signal processors (10)!>), general purpose microprocessors, special application integrated circuits (ASICs), field programmable logic (four) columns (FpGA), or others. Equivalent integration or discrete logic. Accordingly, the term "processor" as used herein may refer to any of the foregoing structures or any other organization suitable for the implementation of the techniques described herein. Further, in some aspects, the functionality described herein Can be provided in dedicated software units or hardware units configured for encoding and decoding, or incorporated in a combined video encoder_decoder (c〇dec). Different features are depicted as units intended to highlight The different functional aspects of the described devices do not necessarily imply that such units must be implemented by separate hardware or software components. In fact, the functionality associated with one or more units can be integrated into common or separate hardware. Various embodiments have been described. These and other embodiments are within the scope of the following claims. [FIG. 1 is a block diagram illustrating a video transmission system that supports video adjustability. 0 is a block diagram showing the source device and destination device of the code writing system of FIG. 1 in further detail. FIG. 3 is a further detailed illustration of an example base layer encoder and enhancement layer Block diagram of the coder. 135453.doc -64 - 200934250 Figure 4 is a block diagram illustrating the example base layer entropy coder and enhancement layer entropy coder in further detail. Figure 5 is a further detailed description of the base layer decoder and enhancement layer A block diagram of an example of a decoder. Figure 6 is a block diagram illustrating an example base layer entropy decoder and a enhancement layer network masher in further detail.Figure 7 is a conceptual diagram illustrating a zigzag scan of a 4x4 coefficient block. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 9 is a flow chart illustrating an example operation of a video encoder embodying the adjustable video coding technique of the present invention. FIG. 10 is a flowchart illustrating an example of operation of a video encoder embodying the adjustable video coding technique of the present invention. A flowchart of an example operation of an enhancement layer encoder that encodes a residual coefficient of a enhancement layer video block. Figure 11 is an example of a enhancement layer decoder illustrating decoding of a enhancement layer bitstream to obtain a vector of residual transform coefficients. Flowchart of Operation ❹ Figures 12 through 15 are block diagrams illustrating different configurations of an encoder and/or decoder for use in an adjustable video dubbing in accordance with the present invention. Element Symbol Description 10 Video Transmission System 12 Source Device 14 Destination Device 14A Multiple Destination Devices 14B Multiple Destination Devices 16 Communication Channels 135453.doc 200934250 ❹ 10 18 Video Source 20 Video Coding 22 Transmitter 24 Receiver 26 Video Decoder 28 Display device 30 Base layer coding Is 32 Enhancement layer encoder 33A Prediction unit 33B Prediction unit 34 Base layer decoder 35A Frame storage 35B Frame storage 36 Enhancement layer decoder 38A Transformation unit 38B Transformation unit 40A Quantization unit 40B quantization unit 41A coefficient scanning unit 41B coefficient scanning unit 42A inverse quantization unit 42B inverse quantization unit 44A inverse transform unit 44B inverse transform unit 135453.doc -66- 200934250

46 Base layer smoke encoder 48 Summer 48A Summer 48B Summer 48C Summer 48X Summer 48E Summer 48F Summer 49 Enhanced Layer Entropy Encoder 50 Analysis Unit 52 VLC Table 52A~52N VLC Table 54 Total Coefficient Encoder 56 Trailing 1 (T1) Encoder 58 Positive and Negative Encoder 60 Series Number Value Encoder 62 and Run Encoder 64 Run Length Encoder 66 ΕΟΒ Symbol Encoder 68 Run Length Encoder 69 VLC Table 70 Positive and negative encoder 72 base layer entropy decoder 74A coefficient scanning unit 135453.doc -67- 200934250 φ 鲁 74Β coefficient scanning unit 76 逆 inverse quantization unit 76 逆 inverse quantization unit 78 逆 inverse transformation unit 78 逆 inverse transformation unit 80 Α prediction unit 80 Β prediction unit 82 Α Frame storage 82 Β frame storage 84 Α summer 84 Β summer 86 enhancement layer entropy decoder 90 total coefficient decoder 92 trailing 1 (Τ 1) decoder 94 positive and negative decoder 96 is a number value decoder 98 and Motion Decoder 100 Run Length Decoder 102 ΕΟΒ Symbol Decoder 104 Run Length Decoder 106 Positive and Negative Decoder 110 Video Block 120 Coefficient Block 170 Example Adjustable Video Encoder 135453.doc 68- 200934250 172 Prediction Unit 173 Frame Storage 174 Transform Unit 175A Quantization Unit 175B Quantization Unit 176A Inverse Quantization Unit 176B Inverse Quantization Unit 177 Inverse Transform Unit φ 178 multiplex module 179 Α summer 179 Β summer 179C summer 180 instance adjustable video decoder 181 demultiplexing module 182 逆 inverse quantization unit 182 逆 inverse quantization unit w 183 inverse transform unit 184 prediction unit • 185 Block storage 186A summer 186B summer 190 instance adjustable video decoder 191 demultiplexing module 192A inverse quantization unit 135453.doc -69 - 200934250 192B inverse quantization unit 193A inverse transform unit 193B inverse transform unit 194 prediction unit 195 Frame Storage 196A Summer 196B Summer 200 Example Video Encoder ❹

135453.doc 70-

Claims (1)

200934250 X. Patent application scope: A method for using an adjustable video write-coded video data method, the method comprising: encoding a video block as a base layer bit stream by a first quality; and 'encoding the video block Refining the at least one enhanced layer bit stream, the refinement causing the video region having a second quality greater than the first quality when combined with the video block encoded with the first quality The block 'where the refinement of the video block is encoded in a single-code one operation. 2' The method of claim 2, wherein the refining encoding the video block comprises encoding each non-zero coefficient of the refinement without analyzing any subsequent coefficients. 3. The method of claiming, wherein the refinement packet 3 encoding the video block encodes & 1 for each non-zero coefficient of the refinement of the video block, exhibiting at least a residual A sign of a non-zero coefficient, a run length indicating the number of other zero-valued coefficients of the non-zero coefficient, and a sign of the non-zero coefficient. 4. The method of claim 3, further comprising encoding a symbol to indicate that there are no residual non-zero coefficients in the refinement of the video block after the last non-zero coefficient. The method of claim 3, further comprising adjusting the magnitude of the non-zero coefficients of the refinement of the video blocks to be equal to one. 6. The method of claim 5, wherein the magnitude of each of the non-zero coefficients such as the 135453.doc 200934250 refined by the video block is equal to one included in the non-coding of the non-coding The coefficients are encoded in the case of magnitudes of zero coefficients. 7', as in the method of claim 1, wherein the refinement encoding the video block as part of the enhancement layer turbulence comprises the refinement encoding the video block, such that the video block is served The equalized coefficients are decodable without accessing the coefficient information of the video block encoded by the first quality as part of the base layer bit stream. 8. The method of claim 1, further comprising encoding the refinement of the video block using only a single variable length code (VLC) table. 9. The method of claim 1, wherein encoding the video block as the portion of the base layer with the first quality comprises encoding the video block with the first quality using a write code technique, the write code technique The first write code analyzes one of the coefficient blocks of the video block in one operation, and encodes the coefficient vector in the second write code one operation based on the analysis. 10. The method of claim 9, wherein: encoding the video block with the first quality comprises using a content adaptive variable according to the ITu_T H.264/MPEG-4 part 1 Advanced Video Write Code (AVC) standard The length code (CAVLC) process encodes the video block with the first quality; and the refinement encoding the video block includes encoding the video area using one of the VLC tables defined in the CAVLC process. The refinement of the blocks. 11. The method of claim 1, wherein the first quality and the second quality comprise one of a first signal to noise ratio (SNR) and a second signal to noise ratio (SNR) and a 135453.doc 200934250 One of a spatial resolution and a second spatial resolution. A device for encoding video data using an adjustable video recording code, the device comprising at least one encoder to: encode a video block as a portion of a base layer bit stream with a first quality, and encode the video block Refining as part of at least one enhancement layer bit stream, the refinement being combined with the video block encoded with the first quality The video block having a second quality greater than the first quality is encoded in the "single-coded one operation" of the video block. α is the apparatus of claim 12, wherein the at least the encoder encodes each non-zero coefficient of the scalar equalization without analyzing any subsequent coefficients. 14. The apparatus of claim 12, wherein the at least-encoder encodes, for each non-zero coefficient of the refinement of the video block, a sign indicating the presence of at least one residual non-zero coefficient, and an indication of the non-zero The length of the run of the number of zero-valued coefficients before the zero coefficient and a sign of the non-zero coefficient. 15. The apparatus of claim 14, wherein the at least one encoder encodes a symbol such that no residual non-zero coefficients are present in the refinement of the video block after encoding a last non-zero coefficient. 16. The apparatus of claim 14, wherein the at least one encoder adjusts the magnitudes of the refined non-zero coefficients of the video blocks to be equal to one. 17. The device of claim 16, wherein the at least one encoder encodes the coefficients without encoding the magnitude of the non-zero coefficients. 18. The device of claim 12, wherein the at least one encoder encodes the refinements of the video zone 135453.doc 200934250, such that the refined coefficients of the video block are not accessed by the first The quality code is decodable in the case of coefficient information of the video block of the base layer bit stream. 19. The device of claim 12, wherein the at least one encoder encodes the refinement of the video block using only a single variable length code (VLC) table. 20. The device of claim 12, wherein the at least one encoder encodes the video block with the first quality using a write code technique, the write code technique analyzing the video block in a first write code operation One of the coefficient vectors, and the coefficient vector is encoded in the second write once operation based on the analysis. 21. The device of claim 20, wherein the at least one encoder: encoding the video block with the first quality comprises using an Advanced Video Recording Code (AVC) standard according to ITU_T H.264/MPEG-4 Part 1 A content adaptive variable length write code (CAVLC) process encodes the video block with the first quality; and the refinement encoding the video block includes using one of the VLC tables defined in the CAVLC process To encode such refinements of the video block. 22. The device of claim 12, wherein the first quality and the second quality comprise one of a first signal to noise ratio (SNR) and a second signal to noise ratio (SNR) and a first spatial resolution And one of the second spatial resolutions. 23. The device of claim 12, wherein the at least one encoder comprises: a base layer encoder 'the base layer encoder encoding the video block as part of a base layer bit stream with the first quality, and The enhancement layer encoder 'the enhancement layer encoder encodes the video block 135453.doc 200934250 refinement as part of the at least-enhancement layer bit stream, the refinement in the video region of the AND, w-th quality encoding When the blocks are combined, the video block having the second quality greater than the first quality is caused. The device of item 12, wherein the device comprises a wireless communication device. 25. The device of claim 12 wherein the device comprises an integrated circuit device. A computer-readable medium containing instructions for causing - or a plurality of processors to: encode a video block as a portion of a base layer bit stream with a first quality; and encode the video region Refining the block as part of at least one enhancement layer bit stream' such refinement results in a video zone having a second quality greater than the first quality when combined with the video block encoded with the first quality A block in which the refinements of the video block are encoded in a single coded one operation. 27. The computer readable medium of claim 26, wherein the instructions cause one or more processors to encode each of the non-zero coefficients of the refinement without analyzing any subsequent coefficients. 28. As claimed in claim 26 Computer-readable medium, wherein the instructions cause one or more processors to encode, for each of the non-zero coefficients of the refinement of the video block, a sign indicating the presence of at least one residual non-zero coefficient, an indication The length of the run of the number of zero-valued coefficients before the non-zero coefficient and a sign of the non-zero coefficient. 29. The computer readable medium of claim 28, wherein the instructions cause one or more processors to encode a symbol to indicate that the encoding is followed by a final non-zero coefficient 135453.doc 200934250 30. 31. 32. 33. 34. 35. There are no residual non-zero coefficients in such refinements of the video block. The computer readable medium of claim 28, wherein the instructions cause one or more processors to adjust the magnitude of the non-zero coefficients refined by the video block to be equal to one. The computer readable medium of claim 30, wherein the instructions cause the one or more processors to encode the coefficients without encoding the magnitude of the non-zero coefficients. The computer readable medium of claim 26, wherein the instructions cause the one or more processors to encode the refinement of the video block such that the refined coefficients of the video block are not accessed The first quality code is decodable in the case of coefficient information of the video block of the base layer bit stream. The computer readable medium of claim 26, wherein the instructions cause the one or more processors to encode the refinement of the video block using only a single variable length code (VLC) table. The computer readable medium of claim 26, wherein the instructions cause one or more processors to encode the video block with the first quality using a write code technique. The write code technique operates once at a first write code A coefficient vector of one of the video blocks is analyzed, and the coefficient vector is compiled in a second write code operation based on the analysis. The computer readable medium of claim 34, wherein the instructions cause the one or more processors to: encode the video block with the first quality comprising 10 according to the ιτυ-τ H-264/MPEG-4 portion The Video Recording Code (AVC) standard encodes the video block with the first quality using a 135453.doc 200934250 Capability Variable Length Write Code (CAVLC) process; and encodes the video block for such refinement The refinement of encoding the video block using one of the VLC tables defined in the CAVLC process is included. The computer readable medium of claim 26, wherein the first quality and the second quality comprise one of a first signal to noise ratio (SNR) and a second signal to noise ratio (SNR) One of the spatial resolution and a second spatial resolution. 37. An apparatus for encoding video data using an adjustable video code, the apparatus comprising: a first component for encoding a video block as a portion of a base layer bit stream with a first quality; and for encoding Refining the video block as a second component of at least one portion of the enhancement layer bitstream, the refinement resulting in having a greater than the first quality when combined with the video block encoded with the first quality The second quality of the video block, wherein the refinement of the video block is encoded in a single coded one operation. 38. The apparatus of claim 37, wherein the second encoding component encodes each non-zero coefficient of the refinements without analyzing any subsequent coefficients. 39. The apparatus of claim 37, wherein the second encoding component encodes for each of the refined non-zero coefficients of the video block - indicating the presence of at least one residual non-zero coefficient symbol, indicating "the non- The length of the run of the number of zero-valued coefficients before the zero coefficient and the sign of the non-zero coefficient. The apparatus of claim 38, wherein the second encoding component encodes a symbol to "not after the encoding - the last non-zero coefficient in the video block (4), there is no residual non-zero coefficient. For example, the equipment of the syllabus 38, the Dan W Di (4) component adjusts the magnitude of the non-zero coefficients refined by the video block to be equal to... 42. If the request item 4 (4), the ten The second encoding means encodes the coefficients without encoding the magnitude of the non-zero coefficients. 43. The device of claim 37, wherein the second encoding component encodes the refinement of the video block such that the (four) coefficients of the video block are not accessed with the first quality code as the basis The coefficient information of the video block of the portion of the layer bit stream is decodable. 44. The device of claim 37, wherein the second encoding component encodes the refinements of the video block using only a single variable length code (VLC) table. 45. The device of claim 37, wherein the first encoding component encodes the video block as the portion of the base layer by using the first code quality to compile the video block with the write quality technique The code writing technique analyzes one of the coefficient vectors of the video block in the first write once operation, and encodes the coefficient vector in the second write once operation based on the analysis. 46. The device of claim 45, wherein the first encoding component performs the following operations: encoding, by the first quality, the video block comprises: 10th progressive video writing code according to the H.264/MPEG-4 portion ( Avc) standard use - content adaptive variable length write code (CAVLC) process to encode the video block with the first quality; and 135453.doc 200934250 encoding the video block number M 1 λ refinement included in the The ones in the VLC table defined in the CAVLC process encode the refinements of the video block. 47. 49. 50. The device of claim 37, wherein the first quality and the second quality comprise one of a first signal to noise ratio (SNR) and a second signal to noise ratio (SNR) Spatial resolution and - in the second spatial resolution. A method for decoding video data using an adjustable video write code, the method comprising: decoding a base layer bitstream to obtain a first quality video block; and decoding the enhancement layer bitstream to obtain the video block Refined, the refinement, when combined with the decoded first video block of the first quality, results in the video block having a second quality, wherein decoding the enhancement layer includes the same for the video block For each non-zero coefficient, a symbol indicating the presence of at least one residual non-zero coefficient, a run length indicating one of the zero-value coefficients before the non-zero coefficient, and a sign of the non-zero coefficient are decoded. The method of claim 48, further comprising decoding, after a final non-zero coefficient, a symbol indicating that there are no residual non-zero coefficients in the refinement of the video block, such as the method of claim 49. Further comprising generating the decoded run of each coefficient, the sign of each coefficient, and the symbol indicating the absence of residual non-zero coefficients to generate a vector of the refined coefficients of the video block. 135453.doc -9-200934250 51. The method of claim 50, further comprising setting a magnitude of each non-zero coefficient of the coefficient vector equal to one. 52. The method of claim 48, wherein the refining the decoding of the video block comprises decoding the video block without accessing coefficient information of the video block encoded with the first quality Refined. 53. The method of claim 48, further comprising decoding the refinement of the video block using only a single variable length code (VLC) table. 54. The method of claim 53, wherein the single VLC table is included in a VLC table as defined in a CAVLC as defined in the ITU-T H.264/MPEG-4 Part 10 Advanced Video Write Code (AVC) standard. One of them. 55. An apparatus for decoding video data using an adjustable video write code, the apparatus comprising at least one decoder for: decoding a base layer bitstream to obtain a first quality video block; and decoding a enhancement layer a bitstream to obtain refinement of the video block, the refinement causing the video block having a second quality when combined with the decoded first video block of the first quality, wherein the at least one decoding Transducing, for each non-zero coefficient of the refinement of the video block, a symbol indicating the presence of at least one residual non-zero coefficient, a length indicating a number of zero-valued coefficients before the non-zero coefficient, and the non-zero A sign of zero coefficient. 56. The apparatus of claim 55, wherein the at least one decoder decodes a symbol indicating that there are no residual non-zero coefficients in the refinement of the video block after a last non-zero coefficient. The apparatus of claim 56, wherein the at least one decoder uses the decoded run of each coefficient, the sign of each coefficient, and the indication that there is no residual non-zero coefficient A symbol is used to generate a vector of the precision coefficients of the video block. 58. The apparatus of claim 57, wherein at least the device sets the magnitude of each non-zero coefficient of the coefficient vector equal to one. 59. The device of claim 55, wherein the at least one decoder does not access the refinement of the video block of the coefficient information of the video block encoded by the first quality. The device of claim 55, wherein the at least one decoder only uses the -single-variable-length write-malay (VLC) table to resolve the refinements of the video block. 61. The device of claim 60, wherein the single VLC table is included in a VLC table as specified in a CAVLC as defined in the ITU-T H.264/MPEG-4 Part 10 Advanced Video Write Code (AVc) standard. One of them. 62. The device of claim 55, wherein the at least one decoder comprises: a base layer decoder that decodes the base layer bitstream with the first quality to obtain the video block; and an enhancement layer decoder that decodes the enhancement layer bitstream to The refinement of the video block is obtained. 63. The device of claim 55, wherein the device comprises a wireless communication device. 64. The device of claim 55, wherein the device comprises an integrated circuit device. 65. A computer readable medium containing instructions for causing one or more processors to: decode a base layer bitstream to obtain a first quality video zone 135453.doc -11- 200934250 Blocking; resolving a code-enhanced layer bitstream to obtain refinement of the video block, the refinement resulting in a second quality when combined with the decoded first video block of the first quality a video block, wherein the instructions cause the one or more processors to decode, for each non-zero coefficient of the refinement of the video block, a symbol indicating that at least one residual non-zero coefficient exists, A run length indicating the number of zero values before the non-zero coefficient and a sign of the non-zero coefficient. 66. The computer readable medium of claim 65, wherein the instructions cause one or more processors to decode after a last non-zero coefficient to indicate that there is no residual non-zero in the refinement of the video block The sign of the coefficient. 67. The computer readable medium of claim 66, wherein the instructions cause the one or more processors to use the decoded run of each coefficient, a 5 cc sign for each coefficient, and indicate that there are no residual non-zero coefficients The symbol is used to generate a vector of the refined coefficients of the video block. 68. The computer readable medium of claim 67, wherein the instructions cause one or more processors to set a magnitude of each non-zero coefficient of the coefficient vector equal to one. 69. The computer readable medium of claim 65, wherein the instructions cause one or more processors to decode the video block without accessing coefficient information of the video block encoded with the first quality These refinements. 70. The computer readable medium of claim 65, wherein the instructions cause one or more processors to decode the refinement of the video block using only a single variable length code (VLC) table. 135453.doc -12-200934250 71. The computer readable medium of claim 70, wherein the single VLC table is included in an Advanced Video Write Code (AVC) standard as defined in ITU-T H.264/MPEG-4 Part 10 One of the VLC tables specified in the CAVLC" 72. A device for decoding video data using an adjustable video write code, the device comprising: for decoding a base layer bitstream to obtain a first quality a first component of a video block; and a first component for decoding an enhancement layer bitstream to obtain refinement of the video block, the refinement being in the decoded video region of the first quality Combining the blocks results in the video block having a second quality, wherein the second decoding component decodes a symbol indicating the presence of at least one residual non-zero coefficient for each of the non-zero coefficients of the refinement of the video block And a run length indicating a number of zero value coefficients before the non-zero coefficient and a sign of the non-zero coefficient. 73. The apparatus of claim 72, wherein the second decoding component decodes a symbol indicating that there are no residual non-zero coefficients in the refinement of the video block after a last non-zero coefficient. 74. The apparatus of claim 73, further comprising: generating, by the decoded path of each coefficient, the sign of each coefficient, and the symbol indicating the absence of residual non-zero coefficients to generate the video block The component of the vector of one of these refined coefficients. 75. The apparatus of claim 74, further comprising means for setting a magnitude of each non-zero coefficient of the coefficient vector equal to one. 76. The device of claim 72, wherein the second decoding component decodes the video block without accessing coefficient information of the video block encoded by the first quality code of the 135453.doc •13-200934250 Equal refinement. 77. The device of claim 72, wherein the second decoding component decodes the refinement of the video block using only a single variable length write code (VLC) table. 78. The device of claim 77, wherein the single VLC table is included in a VLC table as defined in a CAVLC as defined in the ITU-T H.264/MPEG-4 Part 10 Advanced Video Code (AVC) standard. One of them. 135453.doc -14-