TWI355204B - Method and apparatus for decoding/encoding a vide - Google Patents

Description

1355204 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a coding scheme for a video signal. [Prior Art] Generally, compression coding means a series of signal processing techniques for transmitting digital information through a communication circuit or storing digital information in a format suitable for a storage medium. Compressed encoded targets include audio, video, and text. In particular, complete the video

The scheme of compression coding is called video sequence compression. Typically, image sequences are characterized by spatial redundancy and temporal redundancy. In particular, the bitstream of the adjustable video coding can be partially and selectively decoded. For example, a low complexity decoder can decode the base layer, and a low (four) rate bit (4) is captured to pass the limited capacity_road_transmission. In order to gradually produce a highly resolved image, it is necessary to gradually enhance the quality of the image. SUMMARY OF THE INVENTION Accordingly, the present invention is directed to a video conferencing system that avoids the slogan and defects of a conventional technique. Whitening The purpose of the present invention is to provide a efficiencies. The coding of the θ strong coded video signal + the poor reference layer is not the same, the most] the method, the transmission of the information associated with the additional prediction of the region of the enhanced shot. The other purpose is to improve the transmission of the associated information of the inter-layer prediction by confirming the configuration information on the tunable video of the tunable video programming 5 1355204. Another object of the present invention is to provide a method for minimizing the transmission of associated information of inter-layer prediction by confirming information indicating whether or not inter-layer prediction is performed. It is an object of the present invention to provide a method for minimizing the transmission of associated information for inter-layer prediction by confirming quality identification information. Another object of the present invention is to provide a method for enhancing the coding efficiency of video signals by defining information on the processing of the segment boundaries of the segments.

It is a further object of the present invention to provide a method for improving coding efficiency in accordance with the self-set information of a tunable video encoded bit stream that identifies an appropriate location. Accordingly, the present invention provides the following effects or advantages. First, it is checked whether the current block in the enhancement layer can be predicted using inter-layer prediction. If the current block in the enhancement layer is not side-by-side, there is no need to transmit coded information for the inter-layer system. Shame, the present invention improves coding efficiency. Secondly, by identifying the configuration information of the tunable video coding at the appropriate location, the associated transmission resources of the inter-layer prediction are minimized. For example, the axis_ indicates whether the inter-layer prediction is carried out (4) dragon and/or the rhyme of the product, and the correlation transmission of the reward is minimized. The invention can be used to report the processing of the boundary of the segment. Therefore, the method of (4) above can greatly increase the coding efficiency of the video signal. 6 1355204 The person skilled in the art can derive from the invention as appreciated by the present invention. The present invention == generally understands or (d) 昍 the purpose and advantages of the present invention can be transmitted through the present invention. In particular, the brothers and the patent application scope are specifically implemented and obtained. In order to obtain the purpose and other characteristics of the present invention, the present invention is specifically and generally bribed. The decoding of Green (4) includes the following steps.

The bit stream of the first layer; the job information is used to indicate whether the inter-layer prediction 1 is above the current block of the second layer; according to the flag information, at least one difference π 于 于 曰 曰 层 层 预测 预测The difference in position between the increased sampled image of the first layer and the current image of the second layer; and the use of at least one deviation information to increase the reference image of the first layer of the sample. For better navigation, the present is more involved in the transmission, and this information indicates whether the number of samples used to increase the number of samples is present in the corresponding area of the second layer. At least one deviation information is obtained according to the information indicating whether the parameter for increasing the sampling exists in the corresponding region of the second layer. A more preferred case 'This method further includes obtaining information on the phase shift of the color difference signal according to whether the parameter for increasing the sampling exists in the corresponding region of the second layer, wherein the first layer uses the phase shift of the color difference signal The information was added and sampled.曰 In this case, the information on the phase shift of the color difference signal is obtained based on the information indicating the color format. In addition, the information on the phase shift of the color difference signal includes water 7 squared more trr pseudo-shift information and vertical phase shift information. • The first 6 conditions of the 'financial law include the second layer of the _ block of the money. The money is not reduced, used for knowledge. Beixi is based on this quality identification, the second layer and The same, the first layer of silk from the β 4 ratio and the inter-lab resolution is not, ', the same video signal from the first layer of the brother. In order to further obtain the decoding of these objects of the invention « contains: one / He likes - kind of video information 1 匕 3 basic layer ~ Shima early π, decoding the first layer of a header information acquisition unit, after the corpse choose - owe '丨L ' brother X take early π — flag , used to indicate that the current block formed in the second layer between layers, and the sub-segment unit, according to the flag y and V-deviation, is used to indicate the increased sampling image of the first layer used for inter-layer prediction. The positional deviation of the #front image of the second layer; and the addition of the sampling unit, using at least one deviation f to increase the reference image of the first layer of the sample. It will be understood that the generalization of the invention as described and subsequent The detailed description of the invention is a representative and explanatory description, and is intended to be DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The preferred embodiments of the present invention will now be described in detail in conjunction with the drawings. First, the compression coding of video signal data needs to consider spatial redundancy, time redundancy, and Redundancy and inter-view redundancy. The compression coding scheme considering tunable redundancy is an embodiment of the present invention. The technical idea of the present invention can be applied to time redundancy, space 1355204 redundancy, inter-view redundancy, etc. In the disclosure of the invention, the encoding includes two concepts of encoding and decoding. The encoding can be flexibly solved_corresponding to the technical concept and protection scope of the present invention. - Vision-based bit-sequence +, in the Wei coding layer (video) There is a separate layer and structure between the coding .^, VCL and the underlying system. It is called the network extraction layer (take her abStraCti〇n 1; Wei). The video coding layer itself processes the coding of motion pictures (4) The 'bottom system' transmits and stores the encoded information. The output generated by the encoding program is the video encoding layer data and is mapped through the network abstraction layer unit before being transmitted or stored. The network abstraction layer unit contains compressed video data or original byte sequence load (rawbyte, etc.) (original byte sequence load: result data of motion image compression), where the original byte sequence The load is the data corresponding to the header information. The network abstraction layer unit mainly consists of two parts, which are the network abstraction layer header and the original tuple sequence 贞®. The secret extraction track contains the flag# (naLref_idc • The flag f (nal-ref) is used to indicate whether it contains a segment of the reference image that becomes the network abstraction layer unit, and the information (naLunitJype) • indicates the type of the network abstraction layer unit. The compressed initial data Stored in the original bit/tuple sequence payload. The end of the original byte sequence load is added to the end of the original • byte sequence load to represent the length of the original byte sequence load for 8-bit multiplication . The types of network abstraction layer units are instant decoding refresh (leg) image, sequence parameter set (called 9 204 parameter set; SPS), image parameter set (picture parameter milk; pps), supplementary enhancement information (supplementalenhanCementinf_ati 〇n ; sm) and so on. Therefore, if the information used to indicate the network abstraction layer unit type Ual_unit__) can be used to improve the coding performance by adding various configuration information related to the horrible horse. For example, it can increase financial information, dependency identification information, quality identification information, J standard 'r (me^ayU^eu^-flag), priority level, and flag information means whether the current access unit is an instant decoding update. The access unit, the dependency identification information is used to indicate the empty_ya, and the flag information ((10) "coffee-lay hearted" is used to indicate whether or not the interlayer system is used. In order to manage the decoded image buffer more efficiently, various configuration information can be encoded. The following is a detailed explanation with reference to "Fig. 2". = normalization, 'all types and levels need to be set to be able to achieve the appropriate cost. π & In the example, the decoder should meet the requirements determined according to the level. Therefore, define two concepts of type 1, ^ = a function or button, and the secret wire can handle _ _ degrees. Type identifier (p_e idc) The type of knowledge is known. Type Na is based on the specified type. The foundation is based on _ (baselh = as 66 ' tastes the basis of the bit stream is the main _ (she _ 丨〇. If _ recognition = 1355204 shape) means that the basis of the bit stream is extended Type (exiended_)e). In addition, the type identifier can be included in the sequence parameter set. ; , in order to break the phase, you need to type the type of the bit stream, or the type used for the sequence. If you type The bit stream is identified as being used to adjust the order of the sequence, so that at least one additional piece of information is transmitted. This example A, for the type of the adjustable sequence is H. The 264/AVC's additional scheme' indicates the type mode used to handle the videoconferencing. #Because the adjustable video editing *e. shouting g; SVC) is the conventional advanced video, the flat code is added, It is more effective to add the grammar as the additional information than to increase the non-fognitive grammar. For example, when the type identifier of the #discrimination code indicates the type of the tunable code, if the information of the tunable order is increased, the information can be improved. Coding efficiency. The following explains an efficient video signal decoding method. The embodiment shown in Fig. 1 is a block diagram of the tunable video coding system of the present invention. In order to finance the various scenarios and various terminals, the sequence provided to the terminal should also be diversified. If the sequence optimized for each terminal is provided to the corresponding terminal, it means that a sequence source is prepared for the combined value of the various parameters, where these parameters contain the number of frames per second, resolution, per Therefore, the burden of providing an optimized sequence is imposed on the content provider. Therefore, the inner valley k is used to steal the initial sequence of the encoding, making it a compressed sequence data with a high bit rate. When the sequence request is made by the terminal, the content provider decodes the initial sequence, and the sequence data of 1355204 =:=: sequence processing capability, _ for the encoding of four kinds of transcoding (transc〇ding) accompanied by the encoding __ solution The method avoids the supply of the shirt (4) (4). Therefore, the extra hard and complex _ set and algorithm. The other side of the 'adjustable coding code crane-hoc code scheme, suitable for editing ^ Jia Ying H Wei Wei To make the county students' miscellaneous table: over-solution The sequence of codes. In this example, the partial sequence represents the cabinet selected by the order of the two m. The quality of the adjustable video coding can also reduce the image quality of the phase. In this example, the package L The image phase of the 2 inch fluorescent seconds with a small number of frames can be called = Shumei and/or the sequence of relatively large number of frames per second can be called the image of the increased amount of the virtual touch area. The paste realizes the phase representation of the low image product f. However, if the bit rate is lowered, the secret quality is correspondingly degraded. In order to solve the image quality degradation problem, a low bit rate separation auxiliary image sequence can be provided, for example, including small And/or each of the number of shadows. This is the auxiliary layer, and the main image sequence can be called the enhancement layer. 冉 is the basic input = the __ each, the concept package used in the disclosure of the present invention 3 brother - layer and brother second floor. For example, the second layer may comprise a different spatial solution than the first layer. 12 1355204 Resolution or Screen Ratio The second layer may contain a different image quality than the first layer. More specifically exemplified 'the first layer may be a base layer and the second layer may be a reinforcement layer. : Layer • Pre-arrangement, based on the layer can be the reference layer, and the second layer can be the current layer. The basic layers and layers of the following description are merely examples and are not an explanation of the present invention. The following detailed information is about the code. First, the code system includes an encoder 102 and a decoder 110. The encoder 1〇2 & includes a base layer coding unit, an enhancement layer coding unit 106, and a multiplex unit just. The decoder 110 may include a demultiplexing ►K 112, a base layer decoding unit u-cut, and an enhancement layer decoding unit 116. The basic bit stream can be generated by compressing the input sequence signal X(8)' base layer coding unit 1〇4. Using the input sequence signal X(8) and the information just generated by the base layer coding unit, the enhancement layer coding unit 1() 6 is capable of generating an enhancement layer bit stream. Use the base layer bit stream spear: The enhancement layer bit stream, the multiplex unit should be able to produce an adjustable bit stream. The generated turbulent stream is transmitted to the transcoder 11 through the 攸 channel. The tunable bit stream transmitted by the demultiplexing unit 112' of the solution Ma may be separated into an enhancement layer bit stream and a base layer bit stream. The base layer decoding unit m receives the base layer stream ’ stream and then decodes the base layer bit stream into (4) the sequence order of the macro block and the residual and motion lag between the blocks. In this example, the corresponding decoding can be done based on a single loop decoding method. The enhanced Zhan decoding unit U6 receives the enhancement layer bit stream, and refers to the base layer bit stream re-established by the base layer solution 7C 114, and decodes the output sequence signal Xe(n). In this example, the output sequence signal edge (8) is output. Is a low sequence signal for the included image quality 13 1355204 or resolution output signal Xe(n). The enhancement layer decoding unit 116 includes a first header information acquisition unit, a second header credit acquisition unit, a deblocking filter unit, and an increase sampling unit. The first header, the poor acquisition unit, obtains the bit red configuration information from the header of the network extraction layer. For example, the 'first-standard information acquisition unit can obtain quality identification information and is used for the indicator layer.

It is predicted whether there is a Wei-like scale. The second header information obtaining unit can perform the configuration of the bit τ stream from the segment header. For example, the second target acquisition unit can obtain information for instructing the processing of the segment boundary in the sampling program, the 5cL of the mask filter, the information on the phase shift of the color difference signal, and the position difference between the images. Deviation information, whether to implement information on adaptive systems, etc. The deblocking filter unit can solve the associated information of the operation of the block decrement to complete the solution block tear wave. Increasing the sample size allows the position difference between the images predicted by the hybrid phase

The information on the deviation is clearly added to the county-level. Therefore, the information on the first layer of the increased sampling can be used to complete the inter-layer prediction. The 纟' mother-enhancement layer coding unit touch enhancement layer decoding unit 完成 performs coding using inter-time prediction. Lai _ means the information of the ruthenium layer and the sequence of the __ layer. This is the image data or the value of the pixel that belongs to the macro block. For example, the column has an internal basic compensation mode or a residual prediction mode. Internal basic = 杈 "flavors the basic singularity should be, and the style. This example... - tastes 14 z. _ 'Residual pre-requisite ❹ basic layer h contains residual area' is too thin for _ value. Lion (four), Lai The presentation is less. The resolution of the special essay is changed ((10) mplmg), 卩 and 曰 plus Wupsa_mg), etc. For example, the internal sampling can be sampled to complete = pre =. The low sampler is used to generate the sputum data. Can reduce the shadow = degree. This can be called down sampling. In addition, the use of increased waver can generate right-handed pixels #, such as miscellaneous image roughness. Material _ for the increase of two retakes I can include the balance Samples and additional samplings. In this disclosure, 2 sampling can be appropriately explained as follows, the technical concept and the scope of protection of the corresponding embodiments of the present invention. At the same time, the same sequence content is generated for different usages or purposes. The base layer spear strength layers are different from each other in terms of spatial resolution, frame rate, bit rate, etc. Through the layer ____, in the #二的情定U-dyadic) P^f degree of precipitation increases the riding and the basic layer The time is not 2 integral ^ ^ ^ extended spatial scalability ; ESS ) 0 , 4 - When Tian Zeng feed is coded through inter-layer prediction to cover a video signal with a ratio of 16:9 (horizontal vertical), the base layer is encoded as an example with a ratio of 4:3 ^^^. Because the domain layer is encoded in the eiOpping state of the partial knives, even if the basic layer is compensated, the entire area of the enhancement layer is covered. Therefore, because the enhancement layer is 15 1355204 There is no cutting and sampling base layer in the sub-area. In some areas, the basic layer for increasing the interlayer compensation is not used. That is to say, this means that the Lai prediction cannot be applied to the part _. This side scarf, Lin The code information of the inter-layer remuneration will not be transmitted. Reference will be made to "5th drawing", "6th drawing", "7th drawing", "8th drawing", "Fig. 10" and "Digital drawing". It is explained in detail in the examples.

2 and 3 are respectively schematic structural diagrams of an embodiment of the present invention which can be added to the adjustable phase of the adjustable video editing unit and the image of the description configuration information.

"Picture 2" shows an example of the configuration of the network abstraction layer unit on which the configuration of the configuration can be increased. First, the network abstraction layer unit can mainly include 3, 'the route extraction layer unit header and the original byte sequence. The network abstraction layer unit header may include knowledge (4) (naljefjde) and information ((«π%). The identification resource (nal_ref_idc) is used to indicate whether the network abstraction layer unit contains a segment of the reference image, and Information ((nal-unit-type) indicates the type of network abstraction layer unit. The extended area of the network abstraction layer unit header may be limitedly included. For example, if used to refer to the information association of the network abstraction layer unit type The network abstraction layer unit may include an extended area of the network abstraction layer unit header for the adjustable video coding or the instruction to slice the network & layering unit. In particular, if nal-unit-handsome 6 = 2〇 Or 14, the network abstraction layer unit may include an extended area of the network abstraction layer unit header. According to the flag information (svc_mvc_flag) capable of identifying the tunable video coding bit stream, the ancient week sequence may be The configuration information can be added to the extended area of the header of the network abstraction layer unit. 16 The information of the type of the two-way extraction layer unit is the sequence parameter set. , 'If the coffee = 15, then Harahiro The equal-load can include the information on the subset sequence parameter set, the fragment layer (4) the signal preparation solution t, the Lai _: #, the extension area of the sub-cake alcohol number number set. For example, if the type information (pr〇file ^ Sexual video coding _ _, the sub-set scatter can include the phase extension area. Or the 'according to the _ information' sequence parameter set can include the phase parameter set of the application area, the extension of the parameter set can be included in the inter-layer prediction Control the information of the deblocking filter, the characteristics of the filter, the parameters of the information associated with the sampling process, etc. The following explains in detail the various configuration information on the sequence, such as the extended H-domain of the network abstraction layer unit header, and the phase parameter set. The configuration information contained in the extension area and the #层层 towel. First, the flag information (inter_layer_deblocking_filter_control_present_flag) may be obtained from the extended region of the sequence parameter set for indicating whether there is an inter-layer prediction for controlling the deblocking filter. Information about the feature. Information may be obtained from the extended region of the sequence parameter set (extended_spatial_scalability) to indicate the information association of the sampling program. Parameter position. In particular, for example, if extended_spatial_scalability = 0, it means that there are no parameters for increasing the sampling procedure in the sequence parameter set or fragment header. If extended_spatial_scalability = 1, it means that the sequence parameter set exists for 17 1355204 'Parameters for the sampling procedure. If extended-spatial-scalabmty = 2, there is no parameter in the fragment header to increase the sampling procedure. The parameters used to increase the sampling procedure will be explained in detail later with reference to "Section 9". The information 4 for indicating whether or not the inter-layer prediction is used means that the inter-layer measurement is the flag information of the collision when decoding a coded segment. The flag f can be obtained from the extended area of the network abstraction layer header. For example, if the flag information is set to work, it means that the inter-layer prediction is not used. If the flag information is set to 〇, the inter-layer prediction can be used or not used according to the coding scheme in the macro block. This is because the inter-layer prediction in the macro block unit can be used or not used. The enamel identification information 3 indicates information identifying the quality of the network abstraction layer unit. When describing the configuration of the poor news, please refer to "Figure 3". For example, a single image can be encoded as a layer that is qualitatively different from each other. In "Picture 3", layers in Spa_LayerO and Spa Layerl can be encoded as layers that are qualitatively different from each other. In particular, information for identifying the quality of the network abstraction layer unit is named quality_id, layer Bi, B2, B10 can be set to quality_id=0. The layers Q1, Q2, ..., Qi〇 can be set to quality_id=l. That is, layers B1, B2, ..., B10 mean layers containing the lowest image quality. These are called basic images. Layers Q1, Q2, ..., Qi〇 correspond to layers comprising layers Bi B2,..., B10 and contain images of better quality than layers B1, B2, .., B10. Quality identification information can be defined in a number of ways. For example, quality identification information can be expressed as 16 levels. The identification information indicating the spatial adjustability means that when the network extraction layer unit is identified by 1355204 =: when describing the configuration information, please refer to "3rd picture". For example, the dependency 5 varies. "Picture 3" and = over the layer in Spa-UyeH to complete the image obtained by reducing the sticking. In particular, for example, the information identifying the dependency on the network abstraction layer unit is represented as ependency_ld , SpaUyerO t ^ dependencyJd=〇 „ 0

SPa_Uyed t0#jdependeney—Μ1's attack. Dependency identification information can be defined in a variety of ways. Therefore, a network abstraction layer unit containing the same value as the information of the long dependency can be expressed as a dependency representation. At the same time, a single layer can be defined in accordance with the information and quality identification information used to identify dependencies. In this case, the network abstraction layer unit containing the same value as the Weiyi and Weilang information can be expressed as a layer representation. The information used to refer to the adjustability between the sections is expressed as the information of the time grade of the unit of the syllabus extraction layer, and the inter-level can be described in the hierarchical B image structure. For example, the layer of Spa-Layer0 (B1, Q1) and the layer (B3, Q3) may contain the same time level Tem_Layer0. If the layer sleeps a) is the reference layer (Bi, qi) and the layer yang, Q3) 'the egg 5, Q5) may contain the time level Tem-Layer of the layer (B1, Q1) and the layer (Β3, φ). Higher time grade Tem_La_. Similarly, if the layer (B7 shoulder is the reference layer (B1, Q1) and the layer (B5, Q5), then the layer (B7, Q7) can contain a higher level than the layer (B5, (10) __ Tem_Layefl _ level Tem - (four) All network abstraction layer units in a single access unit may contain the same time level value. Instant solution 19 1355204 In the example of the update access unit, the time level value becomes 〇. The flag information is used to indicate the reference basic Whether the image is used as a reference image, the flag information indicates whether the reference county image is used as a reference image in the inter-layer prediction program. The flag information may include the same in the same layer. The value of the network abstraction layer unit, wherein the network abstraction layer unit of the same layer is the network abstraction layer unit that contains the same information of the identification dependency. The priority information service is used for the information extraction layer unit. Priority information. Use the ambiguity information to provide and extend the secret or inter-image scalability. For example, 'Make (10) the first part of the job can provide users with a sequence of multiple and space levels. See Shunqing and Space Or only in accordance with the sequence of different restrictions. Priority information can be formed in a variety of ways according to its reference conditions. Priority information can be randomly formed without special reference, and 'priority information can be determined by the decoder. The configuration information in the extended area of the extracting layer unit header may include flag information, and the flag is used to indicate that the current access unit is a real-time decoding and updating unit. The various information of the inter-layer prediction may be included in the fragment layer command. , may include information for increasing the processing of the segment boundary in the sampling program, 5' associated information with the deblocking filter, information related to the phase shift of the color signal (chrGmasignal), and the position between the indicator layers. Difference_deviation#_, and information indicating the occurrence or non-occurrence of the adaptive 20 13^^204 prediction. The above information can be obtained from the fragment header. Information about the operation associated with the solution block ferrite The example of 6 may be the information indicating the operation of the deblocking filter and the operation method of the wave device C disable-debloeking-fito-ide Required deviation information ( mter-layer-sllce-alpha-cO_ffset_div2, inter_layer_slice_beta_〇ffset_div2), etc. i Yucai> Color information on the phase shift of the information 7 instance, may be used for Inter-layer prediction ~ information on the horizontal and vertical phase offsets of the color distribution (scaled_refJayerJeft_〇fifsets scaled_ref_layer_top_offset, scaled_refJayer_right_offset5 scaled^ref_layer_bottom_offset) 实例 An example of the offset information 8 of the position difference between the private layers may be an indication The interlayer prediction increases the deviation information of the difference between the upper and lower left and right positions between the sampled image and the current image (SCaled_ref_layer_left_offset, scaled.refjayerjop^ffset, scaled1 (layer-right-offset, scaled-ref-layer-bottom_offset). As an example of information indicating that the base layer increases the processing of the macroblocks on the boundary of the sampled tissue segment, it may be the information U〇nstrained_intra_resampling_flag) ' if the corresponding inner coded block of the first layer exists in the first layer more than at least Two segments, the flipping cannot predict the current macroblock by using the corresponding inner coding block in the first layer. 21 : 'United News _ did not show enough rewards "a test brother 11 map" to explain in detail. intention. The picture of the relationship between the base layer and the enhancement layer is not shown in the picture 4 of the sample. It is possible to use the inter-layer prediction when the video is encoded. For example, it is possible to pick up an apricot 3 woody 4 to check whether it is the area corresponding to all the elements in the block; the sub is the basic shot. As for the result of the check, if the current block of the job layer is not used for inter-layer prediction, there is no need to transmit coded information for inter-layer prediction. Therefore, the coding efficiency can be improved. Therefore, a function can be defined that can check that the current block of the enhancement layer can use inter-layer prediction. For example, the function 'in_Cr〇P-Wind〇w〇' can be defined as a function for checking whether the area corresponding to all the elements in the current block exists in the base layer. Assuming that the macroblock index in the horizontal direction on the enhancement layer is set to 'mWdxX, the macroblock index in the vertical direction is set', the function in_cropjwindow() may return the value 'TRUE (or '‘') if the following conditions are satisfied. mbldxX > (ScaledBaseLeftOffset +15)/16 mbldxX < (ScaledBaseLeflOffset + ScaledBase Width -1)/16 mbldxY > (ScaledBaseTopOfifset +15)/16 mbldxY £ (ScaledBaseTopOffset + ScaledBaseHeight _ 1) / Ιό 22 ^55204 'mbldxX' It can be derived using the horizontal macro address and the number of macro blocks. mbldxY can be derived from different methods depending on whether the application of the macroblock adaptive box field is applied. For example, if the macroblock adaptive box field is applied, it can be obtained by considering the macro block pair. When considering a macroblock pair, it is assumed that the index of the upper macroblock is set to 'mbldxYO', and the index of the lower macroblock is set to 'mWdxY1'. ‘mbIdxY〇 can be obtained from the deviation information indicating the difference between the upper portion of the sampled image and the current image used for the inter-layer prediction and the number of macroblocks in the horizontal direction. The value of the 'horizontal macroblock number information' in this example may be different depending on whether the current image is a frame image or a domain image. 'mbldxYi' can be obtained from the information indicating the deviation of the upper position difference between the increased sample image and the current image used for the inter-layer prediction and the macro block number information in the vertical direction. Meanwhile, if the macroblock adaptive frame field is not applied, 'mbldxYO' and 'mbldxYl' can be set to the same value. 4 ScaledBaseLeftOffset, which indicates the deviation information, used to indicate the left position deviation between the increased sample image and the current image used for inter-layer prediction. The 'ScaledBaseTopOffset' indicates that the deviation information' is used to indicate the upper positional deviation between the increased sampled image used for the inter-layer prediction and the current image. ‘ ScaiedBasewidth, which means increasing the horizontal width of the sampled image. ‘ScaledBaseHeight, which indicates increasing the vertical height of the sampled image. The function in_cr〇p_wind〇w() returns the value "FALSE (or '0')' if any of the above conditions are not met. If the 昼23 1355204 element corresponding to at least one element in the current block (CurrMbAddr) is not in the sampling base layer, ie if the function in_crop_window(CurrMbAddr) returns the value 'FALSE' ' then the associated information of the inter-layer prediction is not Used for the current block, this information is not transmitted. Therefore, according to the embodiment of the present invention, if the corresponding base layer area does not exist by in_crop_window (CiirrMbAddr), the transmission of the association information of the inter-layer prediction of the current block can be omitted. An example of using a function in a crop_window to complete a stoner is explained below in accordance with an embodiment of the present invention. First, if it is recognized by ‘in_crop_wmdow(CurrMbAddr)' that the area corresponding to the current block exists in the base layer, the enhancement layer coding unit 1〇6 performs inter-layer prediction using the material and/or motion information of the basic layer. In this example, the motion information may include reference index information, motion vector information, segmentation information, and the like. If the material and/or motion information of the current block is set to the material and/or motion information of the corresponding block, or the material and/or motion information of the current block is obtained from the material and/or motion information of the corresponding block. Then, the enhancement layer coding unit 1〇6 adds the instruction information or the obtained information to the enhancement layer data stream, and then notifies the decoder 11 of the additional information. However, if it is recognized by %-emP_window(CurrMbAddr)' that the area corresponding to the current block does not exist in the base layer, the enhancement layer coding unit can generate the enhancement layer without performing inter-layer prediction. Meanwhile, if the decoder 110 confirms that the area corresponding to the current block does not exist in the base layer through m-CIOP_wmdow (CurrMbAddr)', the riding code n 11G determines that the command information is not transmitted. 24 and FIG. 6 are respectively schematic diagrams showing the macroblock and sub-macroblocks of the inter-layer prediction according to the embodiment of the present invention. The information related to the 4th segment of the current clear extraction layer and the inter-layer prediction is transmitted to the solution. For example, in the case where the enhancement layer predicts the motion vector of the secret block, the flag indicating whether or not to use the motion vector of the base layer (m〇ti〇n_predicti〇n_flag_1χ) can be obtained from the macroblock layer. In accordance with an embodiment of the present invention, in accordance with the manner of checking ‘in_cr〇p_wind〇w(Curr]V[bAddr), the decoder can know whether the information about the inter-layer prediction is transmitted by the encoder [51〇, 61〇]. For example, according to 'in_Crop_wind〇w(CUrrMbAddr)' 'if the area corresponding to the current block does not exist in the base layer, the flag 'm〇ti〇n_prediCti〇n_flag_10/11 is not transmitted on the bit stream [520 /530, 620/630]. The flag 'adaptive_motion_prediction_flag' indicates whether information associated with motion vector prediction occurring within the macroblock layer can be obtained from the segment data of the current network abstraction layer. In accordance with an embodiment of the present invention, information associated with inter-layer prediction is not transmitted by the encoder [510] in accordance with the check 4 adaptive_motion_j)rediction_flag, and ‘in_crop_window(CurrMbAddr)'. For example, according to 'in_crop_window(CurrMbAddr)', if the area corresponding to the current block does not exist in the base layer, or according to 'adaptive_motion_prediction_flag', if the information associated with the motion vector prediction does not exist within the macro block, the flag 'motion_prediction_flag_l0/ 1Γ was not transmitted [520/530, 620/630]. The above technical concept can also be applied to the 25 1355204 sub-macroblock prediction shown in Figure 6. After identifying two types of beacons, only two conditions are met, and the information associated with the layers is transmitted. Therefore, the coding efficiency can be improved. 7F of Fig. 7 is a schematic diagram of the relevant syntax of the residual prediction through the measurement of the invention. When the inter-layer remuneration is completed, the information associated with the # segment data towel of the current network abstraction layer and the inter-layer prediction is transmitted to the decoding ^. For example, when the #_block_remaining signal is measured, the flag ‘residuaLprediction_flag’ used to indicate whether the residual signal of the base layer is used can be obtained from the macroblock layer [74〇]. In this example, the base layer is not known using the layer table. According to an embodiment of the present invention, the information associated with the inner layer prediction is not transmitted by the encoder by confirming the manner of 'in_CmP_wind〇W(CUrrMbAddr)'. For example, depending on the adaptive_residual_prediction_flag, and the fragment type of the current block, a 'residual_prediction_flag' is obtained, wherein the information 'adaptive_residual_prediction_flag' indicates the occurrence of the associated information of the prediction of the residual signal in the macroblock [710]. The 'residual_prediction_flag' can also be obtained according to £base_mode_flag'. ‘base_mode_flag, indicating whether the type (mb_type) of the current macroblock is derived from the corresponding area of the base layer [720]. The 'residualjprediction_flag' can also be obtained according to the type of the current macroblock and the function in_crop_window(CurrMbAddr). For example, when the types of macroblocks and sub-macroblocks are not internal mode [MbPartPredType(mb_type, 0) != Intra_16xl6(8x8 and 26 1355204 4x4)], and the value of in_crop_wind〇w(CurrMbAddr) is 'plus 6, When it means that the corresponding area of the current macro block exists with the base layer, you can get 'iresiduaLprediction Jlag' [730]. If the current macroblock type is not the internal mode' or the corresponding region of the current macroblock does not exist in the base layer fm_〇Op_windGw(CunMbAddr) = 0], the residual prediction is not completed. Encoder 102 generates an enhancement layer 'but does not include 'residuaijjredicti〇n_flag'. If ‘residual_j)rediction_flag is set to '1, the residual signal of the current block is predicted in the residual signal of the base layer. If 'residuai_predicti〇n_flag, the residual signal set to '0' ' is encoded without inter-layer prediction. If "residual_prediction_flag" does not exist in the macroblock layer, then follow the method below. For example, 'residuai_prediction fjag, is paid as a default value (default_residual_prediction_flag) only if the following conditions are fully satisfied. First, 'base_mode-flag' should be set to '1', or the type of the current macroblock should not be an internal core. Next ‘in a crop_window(CmTMbAddr)’ should be set to 1 . The flag 'no-inter-layer_pred-flag' indicating whether the inter-layer prediction is used should be set to '〇'. Fourth, the fragment type should not be an EI fragment (ie, an I fragment or a Si (Switching I) fragment). Otherwise it can be ‘0,. When the corresponding region of the current sequence block does not exist in the base layer by 4in_crop_window(CurrMbAddr), the enhancement layer decoding unit 116 determines that the motion prediction flag (motion_prediction_flag) information does not exist in the macroblock or the sub-macroblock, and 27 1355204 and re-establishing the video using only the (four)-bit stream of the (four) layer without the inter-layer hierarchy. If the residual syntax element is not included in the bitstream of the enhancement layer, the enhancement 1 decoding unit 116 can obtain the __level eSidUal_^edietiQn_flag, 16#in_eiOp_windc)w (CurrMbAddir considers whether the corresponding area of the secret block exists in the base layer t. If K-windGw (CmTMbAddr) is set to 'G', the enhancement layer decoding unit 116 can confirm ## The corresponding area of the sequence block does not exist in the base layer, Residualjrediction.flag^i#*^ '〇', and then the video signal can be re-established using only the information of the enhancement layer without using the residual signal of the base layer. Residual prediction. Figure 8 is a grammatical structure diagram for performing deblocking filtering in accordance with the occurrence or non-occurrence of inter-layer prediction execution according to an embodiment of the present invention. First, in accordance with an embodiment of the present invention, the encoder does not transmit the associated information of the inter-layer prediction by confirming the configuration information of the tunable video encoded bit stream. The configuration information of the tunable video encoded bit stream can be extracted from the network header. Obtained in the extended area. For example, according to the poor information 'n〇-hter layer_pred flag', and the quality identification information (quality_id), the information about the deblocking filtering is obtained [81〇]. For example of the associated information of the block filter operation, there may be a deviation information (disable_deblocking filter idc) used to indicate the solution block 遽波( 木 木 & ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( s s s s s s s s s Div2 — , slice_beta_offset_div2), etc. First, according to the information used to control the characteristics of the deblocking filter, the operation information of the block filter can be obtained by using 28 1355204. In this example, as shown in Figure 2 No, information for controlling the characteristics of the deblocking filter can be obtained from the sequence parameter set. For example, as for controlling the deblocking filter. Of the information, there may be not only the flag information (inter_layer_deblocking a fllter_control_present flag) 'is used to indicate whether the information (820) for inter-layer prediction of the deblocking filter control feature of the present. Thus, information for indicating the method of operation of the deblocking filter can be obtained in accordance with the flag information (830). In particular, if disable_deblocking_filter_idc=0, the filtering can be done over the edges of all blocks of the smear and chromatic aberration signals. If disable_deblocking a filter_idc = 1, the filtering is not completed above the edge of all blocks of the current image. If the disable_deblocking_filter_idc = 2 ′ filter, the wave can be completed above all the edge of the block except for the overlapping segment boundaries. If disable_deblocking_filter_idc = 3, filtering can be done over the edge of the block containing the non-overlapping segment boundaries, and then filtering is done over the block boundary containing the overlapping segment boundaries. If disable_deblocking_filter_idc = 4, the filtering is only done above the block boundary of the luminance signal and not above the block boundary of the color signal. If disable_deblocking_filter_idc = 5, the filtering is done in the luminance signal above the edge of all the blocks above the edge of the block containing the overlapping segment boundaries, above the edge of the block that is not completed by the color signal. If disable_deblocking_filter_idc = 6, the filtering is not done above the edge of the block of the color signal, but only over the edge of the block of the luminance signal. After filtering is done over the block boundary containing the luminance of the un-overlapping segment boundary, the filtering can be done over the block boundary of the luminance signal containing the overlapping segment boundaries. Based on the information used to indicate the operation of the deblocking filter, the offset information required to deblock the block filter can be obtained. For example, if disable_debl〇ddng_mter_idc =1, the deblocking filtering is not completed above the edge of all blocks. Therefore, the deviation information required to solve the block 遽波益 can be obtained only when the value of 'disable_deblocking_filter_idc' is not set to 1. E.g,

‘inter'ayer_slice_alpha c0_offset div2, _ and ‘inter·-layer_Slice_beta_0ffset_div2' indicate the deviation information (85〇) used by the deblocking filter table in the access macroblock in inter-layer prediction. Therefore, the deblocking filtering can be done using the obtained deviation information. Fig. 9 is a diagram showing the syntax structure for obtaining the deviation information indicating the positional difference between the enhanced sampled reference image and the current image in accordance with the inter-layer prediction execution in the embodiment of the present invention. First, in accordance with an embodiment of the present invention, the encoder does not transmit the inter-layer prediction value associated information by checking the adjustable video coding bit stream value configuration information. The configuration information of the tunable video encoded bit stream can be obtained from the extended area of the network abstraction layer header. For example, according to the secondary=negative (njnterjayet^pred_flag) and quality identification information (quality_id) used for whether the upper layer prediction is used, the parameter related information of the sampling program may be increased (910). As for adding spears for sampling. Examples of sequence related information may have a phase shift of the color signal __ Bei Xun 1355204 (930), deviation information indicating the position difference between images (940), and the like. The information related to the parameters used to increase the sampling procedure can be obtained from the extended region of the sequence parameter set and the segment header. As for the information on the phase shift of the color signal (930), there may be information on the horizontal phase shift of the color component of the image used for inter-layer prediction (ref-layer-Chr〇ma_phaSe-X_plusl), and Information on vertical phase change ' ( ref-layer-chroma-Phase_y_Plusl ). As an example of the deviation information (940) indicating the difference in position between images, there is deviation information for indicating the deviation of the up and down and left and right positions between the increased sample image and the current image used for inter-layer prediction (scaled__refJayerJeft_offset, scaled_refJayer_t〇p_〇ffSet5 scaled —ref_layer_right—offset, scaled ref—layer—bottom—offset). According to the information indicating the position of the parameter related information for increasing the sampling procedure, C extension-SpatiaLsealabmty)' is available for increasing the parameters of the sample. For example, if extended scalab^ is set to 〇, it means for & The information about the number of teas added to the autumn order is neither in the sequence parameter set nor in the slice. If the time is made - the printing coffee - (four) secret is set to be U. The parameter information used to increase the sampling procedure does not exist in the segment header and is stored in the phase parameter set. If e_ded_sp-% is set to be good, the parameters related to the parameters that are not used to increase the sampling procedure are not present in the sequence parameters. Therefore, if the extended-sPatial-scalability, 疋 is within the 2 (four) # segment system _ to increase the sampling secret parameter phase 31 1355204 off information (920). In addition, if the extended-talk (four)-(four) coffee service is set to ^, the control sequence parameter set is used to increase the parameter related information of the sampling program. The information on her offset (93G) of the color signal and the deviation information (940) indicating the difference in position between the images can also be used to increase the sampling procedure. Fig. 10 is a grammatical block diagram showing the flag information for the use of the inner block of the base handle according to the presence or absence of the inter-layer prediction execution according to the embodiment of the present invention. First, in accordance with an embodiment of the present invention, by checking the configuration information of the tunable video coded bit stream, the code associated with the 11* wei layer is encoded. The coded address can be summarized. ^: The configuration information of the stream can be extracted from the network. For example, according to information indicating whether or not the inter-layer prediction is used (n〇-imerJayer), and quality identification information (qualityjd), information for processing the segment boundary in the sampling program is increased (1010). As for the information instance for indicating the processing of the segment boundary, there is information (_trained-intra_resampling_flag) for indicating whether or not to restrict the use of the second layer towel for the first layer (10) block of the current block (deletion). The decoding speed at the time of performing parallel processing can be improved by defining information for indicating whether or not to limit the use of internal blocks. Information used to indicate whether to restrict the use of internal blocks is available from the fragment header. Because the information used to refer to whether 7F is restricted to the use of the block can be obtained from the slice header 'even if its value is set A 1 ' needs to check whether the reference block in the first layer corresponding to the current block contains In a specific segment of the layer. Therefore, it can be confirmed that when 32 1355204 constrained should be set _d-^ is set to whether the first-shot reference block is included in the first layer of the second=pre-block pair, if the reference-block in the tier-layer overlaps the first layer In the middle to x, for example, _ is recorded as the first-shot un-internal cipher. Pull segment, current block Basic prediction mode coded as energy block. Internal W uses the internal red equivalent of the corresponding sampling pedal layer. The block coded in the internal mode. If the second layer in the first layer is included in the specific slice of the slate layer, the inner block of the first layer can be used to decode the "block. In the case of the job, the current block can be marked as used (10) basic prediction mode 0 if constrained_intm_resampiing_flag It is set to }, and the information (diSable_debl〇cking_filter-idc) of the operation method of the deblocking filter described in conjunction with "Fig. 8" is limited. For example, disable-deblocking—filter idc can only be set to Bu 2 or 4. If constrained_intra_resampling_flag is set to 〇, even if the corresponding block in the first layer overlaps at least two of the first layer, the inner block of the first layer can be used to decode the current block of the second layer. The above embodiment can be applied to a chr〇 minance signal and a luminance signal in the same manner. Fig. 11 is a diagram showing the syntax structure for obtaining adaptive prediction information in accordance with the occurrence or non-occurrence of inter-layer prediction according to an embodiment of the present invention. 33 1^55204 In accordance with an embodiment of the present invention, the _ information of the inter-layer side of the configuration information by confirming the tunable video encoded bit stream is not read from the encoder. The tunable video encoding bit 'meta-stream configuration information can be obtained from the extended area of the network S layer header. For example, according to the information used to indicate whether the inter-layer prediction is used ίnoJnter_layer_J?red_flag^ > ttfL[ln〇] 0 • The system f can indicate that the heterogeneous W syntax exists in the green setting. For example, there is a tm^adaptive_prediction_flag^ > ^ information that exists in the macro block layer in the segment header and the macro block layer. ^^^^^^, · Information indicating whether residual prediction information exists in the macroblock layer 'adaptive_residual_prediction_flag, and so on. If the inter-layer remuneration is used to indicate whether the inter-layer paste is completed, the flag information used to indicate the presence or absence of the segment data is ‘(4) may be first to [112G]. By confirming the information indicating the hot occurrence of the segment, it can be determined whether or not the inter-layer test is completed within the macro block. According to the indication fragment 'The information of the dragon appears, if the fragment data exists in the fragment _30], the adaptive prediction flag ‘adaptive』redictionJlag, [114〇] can be obtained. Information is also available, ‘adaPtive^_』__n—flag’ ’ indicates whether the confusing association is present in the macroblock layer [_. According to the adaptive prediction flag, the information default_base_mode_flag ’ can be obtained from the basic layer to obtain information indicating whether to delete motion information [(10)]. If the motion # signal is not predicted from the corresponding block of the 34 1355204 base layer [1155], the information 4adaptive_motion_prediction_flag' can be obtained to indicate the related information of the motion prediction. • Whether it exists in the macroblock layer [1160] in. If the motion prediction associated information is present in the macroblock layer [1165] ', the information 'default_motion__predicti〇n flag' can be obtained, indicating how to infer the motion prediction flag information [1170]. * The information 'adaptive-motion_prediction_flag, and the information - 'adaptive_residual_prediction_flag' are used in the macroblock layer, where the information • 'adaptive-motion-Prediction-flaS' is used to indicate whether the associated syntax of the motion prediction exists in the macroblock layer. The information 'adaptive-residual_predicti〇nJlag' is used to indicate whether the associated syntax of the residual prediction exists in the macroblock layer. For example, according to 'adaptive motionprediction flae -sr π " - -s 4 service to the flag 'm〇ti〇IU) rediction_flag_lx' ' is used to indicate whether to adapt to the motion vector of the base layer. According to 'adaptive-residUal_predicti〇n- Flag, the flag _flag' can be obtained to indicate whether it is adapted to the residual information of the base layer. As described above, the decoder/code of the present invention can be provided to the broadcaster/receiver of the multimedia broadcast. In the solution of the signal, data signal, etc. - Multimedia __ can be digital multimedia (four) (Na 丨 如 • — — — — — — — — — — — 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 The decoding/encoding method applied by the invention is configured with a computer-executed program, and the record age stored in the electro-radio reading after missing is included in the invention: #料结构的35 /, - data can be stored in a computer readable In the recording medium, the computer readable recording medium includes various storage devices for storing data that can be read by the computer system. The battery that can be used for drinking 34 includes read-only memory, random access memory, The disc only has a memory, a tape, a floppy disk, an optical data storage device, etc., and also includes a device implemented by (i, such as k over the Internet transmission). The bit stream generated by the encoding method is stored in The invention is transmitted in a computer-readable recording medium or through a line/wireless communication network. - Although the invention has been disclosed above in the preferred embodiments of the foregoing, it is not intended to limit the invention. The modifications and refinements of the present invention are within the scope of the invention, and are not subject to the spirit and scope of the invention disclosed in the appended claims. For the scope, please refer to the attached patent application. [Simplified Schematic] FIG. 1 is a block diagram of a video encoding system of the present invention; FIGS. 2 and 3 are respectively shown in the embodiment of the present invention. It can be added to the configuration of the tunable video coded bit stream and the structure diagram for describing the image of the configuration f; 1 $ Fig. 4 is a schematic diagram of the relationship between the sampled base layer and the enhancement layer ; Fig. 5 and Fig. 6 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Corresponding syntax diagram; Figure 8 shows the syntax structure diagram of the solution decoding according to the embodiment of the present invention according to the occurrence or non-occurrence of inter-layer prediction execution; FIG. 9 is a diagram showing the inter-layer prediction according to the embodiment of the present invention. The grammatical structure diagram of the deviation information of the position difference between the image and the current image is obtained.

Figure 10 is a grammatical structure diagram of the flag-like news of the use of the rewards of the rewards of the present invention for the existence or non-existence of the rewards of the new layer (4); 11 Figure * is the actual implementation of the invention according to the emergence of the foot machine in the county and Lin Qian __ Huan silk knot. I have not [main element symbol description] 102 encoder 104 base layer coding unit 106 enhancement layer coding unit 108 multiplex unit 110 eliminator 112 demultiplexing unit 114 base layer decoding unit 116 enhancement layer decoding unit 37

Claims (1)

X. Patent application scope: 1. The decoding method of the video signal includes: \ Decoding a bit stream of the first layer; : Obtaining - flag information, indicating that the inter-layer prediction is completed in the second layer _ above the current block; " based on the flag information to obtain at least one deviation information, used to indicate between the layer, the first layer of the paste - the increased sample image and one of the second layer of the current image A positional difference; and the field uses at least the one deviation information to increase the sampling of the first layer - the reference image. 2. The decoding method of the video signal as described in the application scope of the patent scope, further includes obtaining a lag message, and indicating whether the parameter is added to the corresponding area of the second layer, wherein at least the The deviation f is obtained according to whether the parameter indicating the increase of the sampling is present in the riding layer of the second layer. • 3. For the decoding method of the video signal as described in item 2 of the patent application, it further includes the root, Wei (4) increasing the number of the mosquitoes in the corresponding area of the second layer of the thief and 4 to the color difference signal. A piece of information on a phase offset, wherein the first layer is increased in sampling using the information on the phase offset of the color difference signal. 4. The method for decoding a video signal according to claim 3, wherein the information on the phase offset of the color difference signal comprises a phase shift information in a horizontal direction and a phase in a vertical direction. Offset information. 5. The method for decoding a video signal as described in claim 2, further comprising: 38 quality identifying the item _: a layer of the current block of the layer, wherein the deviation information is based on the quality identification information And get it. 6. If you apply for a patent range! In the decoding method of the video signal, the middle layer of the basin is the same as the first layer in the screen analysis technique, and the first layer is from the same video signal of the second layer. 7. The decoding device of the video signal comprises: - a base layer decoding unit, a depletion-first layer-bit stream; a first-header information obtaining unit, and the lining indicates whether the inter-layer prediction is completed in a second One of the layers above the current block; - the second header Wei acquisition unit, according to Wei Biao Information Lin, at least one deviation ^ is used to indicate the first layer of the inter-layer prediction - increasing the sample image and the second a positional difference between the current images of one of the layers; and - increasing the sampling unit, using at least the one of the deviation information to increase sampling of the reference image of the first layer. κ 8. The method of decoding a video signal as described in claim 1, wherein the video § is received as a broadcast signal. 9. The method of decoding a video signal as described in claim 1 wherein the spoofing message is received through a digital medium. 10. A computer readable medium for recording a program for decoding a video signal as described in claim 1 of the patent application. 39