JP2016213698A - Image coding device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image coding device and a program which, in the case of compression coding and multiplexing images of plural resolutions, is capable of transmitting the images within a band of a transmission path without deterioration of the images.SOLUTION: A low resolution image and a high resolution image are respectively coded and when a generated coded information quantity exceeds a predetermined value, the phase of processing units in a time (frame) direction of GOP (Group Of Picture) is deviated between the low resolution image and the resolution image, thereby smoothing a generated information quantity generated at the same time and suppressing the total generated information quantity of the images within a bandwidth.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a video encoding device, and more particularly, to a device and a program for compressing and multiplexing two or more types of spatial resolution video.

  In recent years, in the broadcasting field, development and application of ultra-high-definition video such as 8K (spatial resolution 7680 × 4320) and 4K (spatial resolution 3840 × 2160) are progressing. However, for broadcast (distributed) video, it may not be sufficient to use ultra high-definition video alone, depending on the status of the receiving device (broadcast receiver, TV reception function, display resolution, etc.) It is assumed that the same content is required to be transmitted with a plurality of resolutions, such as simultaneous broadcasting of 8K and 4K as well as broadcasting of 8K alone, and broadcasting including 2K.

  When transmitting such ultra-high-definition video, highly efficient compression coding technology is indispensable. Video compression and coding technologies such as MPEG (Moving Picture Experts Group) -2, AVC (Advanced Video Coding) /H.264, and HEVC (High Efficiency Video Coding) /H.265 have been developed, standardized, and used. ing. At the time of transmission, such a compression encoding method is appropriately selected and used.

  In order to simultaneously transmit a plurality of videos, each video signal is compressed and encoded, multiplexed, and transmitted as a multiplexed stream to a transmission path. However, since there are restrictions on the bandwidth (bit rate) and the like in the usable transmission path, efficient compression coding and multiplexing that can effectively use the bandwidth are required.

  When the image quality is kept constant by compression encoding, the amount of information generated varies depending on the video content / picture to be compression encoded. This method is called VBR (Variable Bit Rate). On the other hand, when the bandwidth of the transmission path is fixed such as broadcasting, the generated information amount is controlled so as to keep the generated information amount constant. This method is called CBR (Constant Bit Rate). Furthermore, there is a technique for performing control to keep the total amount of information generated for each content constant when transmitting a plurality of contents while the bandwidth of the transmission path is limited. This is called statistical multiplex coding, and is used to use a band as efficiently as possible when maintaining the highest possible image quality in a narrow band.

  In addition, for the transmission of the same content at a plurality of resolutions, a hierarchical encoding technique has been developed as an extension of a normal compression encoding system. This is because a plurality of (for example, two types) spatial resolution images are compressed and encoded as a single stream, and only a part of the video is decoded when only a part is decoded on the decoding side. This is a compression encoding method having a function of obtaining a high video.

  On the other hand, the transmission band of terrestrial and satellite waves, which are existing broadcasting systems, is limited by channel plans established in the world and due to technical limitations, and transmits ultra-high definition video with multiple spatial resolutions. There may be cases where sufficient bandwidth cannot be secured. In this case, the total amount of information generated by compression encoding of each video instantaneously exceeds the amount of information (bit rate) that can be transmitted, and as a result, there is a risk that the video will be deteriorated or corrupted.

  In response to such a problem, a multi-channel encoding device has been proposed in which the amount of target generated information after multiplexing is made constant by shifting the phase of a GOP (Group Of Picture) of a video signal between channels. (Patent Document 1).

  In the multi-channel encoding device described in Patent Document 1, the encoding of the I picture (intra picture) of each channel is dispersed on the time axis, and the multiplexed stream may exceed a predetermined code rate. Is detected, the I picture prohibition signal is output, and the structure determination unit that has received the I picture prohibition signal prohibits intra-frame encoding by all the image encoding units and performs predictive encoding other than intra-frame encoding. It is configured.

Japanese Patent No. 3147859

  As described above, when a plurality of ultra-high-definition videos are compressed and transmitted in a band that is not sufficient for transmitting the ultra-high-definition video, the decoded video may be greatly deteriorated or broken.

  In particular, in a broadcast system, it is assumed that the same content is transmitted at different resolutions, and since the same content, the amount of information in each frame of the video matches, and the total amount of information generated instantaneously by compression coding is The probability of exceeding the bandwidth of the transmission path is higher.

  When the encoding device of Patent Document 1 detects that a multiplexed stream may exceed a predetermined code rate, the encoding device prohibits I picture encoding, while the I picture prohibition signal is being output. Since the I picture disappears in all the channels, the I picture decreases as a whole, and there is a possibility that the image deteriorates. Further, in Patent Document 1, normally, the arrangement of I pictures is simply shifted in a sequential manner, and an optimal I picture arrangement method is not disclosed. Furthermore, the case where the same content is distributed at different resolutions is not studied, and the application to hierarchical coding is not considered.

  In addition, as a conventional video encoding device, there is a device that prepares a buffer that summarizes information of a plurality of frames and performs encoding within the band of the transmission path. However, the quantization is controlled according to the buffer capacity. Therefore, the amount of generated information is reduced, and the amount of information is flattened in a certain time section. Therefore, there is a possibility that image quality is deteriorated and processing time is delayed.

  Accordingly, an object of the present invention, which has been made in view of the above problems, is to perform video transmission within the bandwidth of a transmission line without degrading video when a plurality of videos are compressed and encoded and multiplexed. An object is to provide a video encoding device and a program which can be made possible.

  In order to solve the above problems, a video encoding device according to the present invention is a video encoding device that simultaneously compresses and multiplexes a plurality of different videos, and when the amount of encoded information generated exceeds a predetermined value. Changing the phase of the reference structure for motion estimation / compensation when encoding at least one of the plurality of different videos, and smoothing the amount of encoded information generated at the same time to the predetermined value or less. It is characterized by.

  The video encoding device multiplexes a plurality of video encoding units and outputs of the plurality of video encoding units, and the amount of encoded information generated by the plurality of video encoding units is the predetermined value. And a phase of a reference structure for motion estimation / compensation at the time of encoding of at least one video encoding unit of the plurality of video encoding units based on the command It is desirable to include a GOP control unit to be changed.

  In the video encoding device, the plurality of video encoding units include a first video encoding unit that encodes a high-resolution video and a second video encoding unit that encodes a low-resolution video. Preferably, the input video signal to the second video encoding unit is a video signal obtained by down-converting the input video signal to the first video encoding unit.

  In order to solve the above problems, a video encoding apparatus according to the present invention is a video encoding apparatus that simultaneously compresses and multiplexes a plurality of videos having different resolutions, and the compression encoding is basically performed. In a video encoding device that is a compression encoding having a structure for predicting a high-resolution video from a video, when the amount of encoded information generated exceeds a predetermined value, at least one of the plurality of video with different resolutions The phase of the motion estimation / compensation reference structure at the time of video encoding is changed, and the amount of encoded information generated at the same time is smoothed below the predetermined value.

  The video encoding device includes a video encoding unit that encodes a basic video, a plurality of video encoding units including a video encoding unit that encodes a high-resolution video, and the basic video. Decodes the output of the video encoding unit to be encoded to generate a local decoding unit that generates a low-resolution video, and outputs the low-resolution video to a video encoding unit that encodes the high-resolution video Multiplexing the output of the plurality of video encoding units and outputting a command when the amount of encoded information generated by the plurality of video encoding units exceeds the predetermined value And a GOP control unit that changes a phase of a motion estimation / compensation reference structure at the time of encoding of at least one video encoding unit of the plurality of video encoding units based on the command. Is desirable.

  Further, the video encoding device has a reference structure for motion estimation / compensation at the time of encoding at least one video of the plurality of video encoding units when the amount of generated encoded information exceeds a predetermined value. It is determined whether the amount of encoded information generated by shifting the phase by a predetermined number of frames is within the predetermined value. If it is within the predetermined value, the reference structure is set to the shifted phase, and if it exceeds the predetermined value, it is further predetermined. It is desirable to repeat shifting the number of frames.

  In order to solve the above problems, a program according to the present invention causes a computer to function as the video encoding device.

  According to the video encoding device of the present invention, when a plurality of ultra-high-definition videos are compressed and transmitted in a band that is not sufficient for transmitting the ultra-high-definition video, the video is not deteriorated (intra pictures are transmitted). It is possible to smooth the instantaneous total generated code amount and enable transmission within the band without preventing the image quality from being deteriorated or broken. In addition, since the processing time of the video itself does not change, there is no difference in delay between a plurality of videos.

  Furthermore, according to the present invention, when hierarchical coding is performed, the picture quality of the enhancement layer can be improved by effectively utilizing the I picture of the base layer in the enhancement layer coding.

1 is a block diagram of a video encoding device according to Embodiment 1. FIG. It is a figure which shows the GOP structure and information generation amount of the conventional two images. FIG. 6 is a diagram illustrating a GOP structure and information generation amount of two images according to the first embodiment. It is a flowchart which controls the phase of GOP structure. 6 is a block diagram of a video encoding device according to Embodiment 2. FIG. It is a figure which shows the GOP structure and reference relationship of the conventional hierarchy coding. FIG. 10 is a diagram illustrating a hierarchical encoding GOP structure and a reference relationship according to the second embodiment.

  Embodiments of the present invention will be described below. In the following description, an input signal as a high resolution video will be described assuming an ultra high definition video such as 8K or 4K, but the high resolution video is not limited to 8K or 4K. In order to simplify the description, it is assumed here that the same video (content) of two different resolutions is transmitted, but the present invention is not limited to two types of different resolution video.

There are the following three methods for transmitting while keeping the total encoded information amount of the two types of images constant while the transmission band is limited.
(1) A method of independently CBR encoding each video and maintaining a constant bit rate (2) A method of maintaining a constant bit rate by statistical multiplexing of two types of video (3) After using hierarchical encoding , How to keep a constant bit rate

  The method (1) applies normal CBR encoding to each video, and there is no correlation between information between the two videos, and the total generated code amount is actively reduced or adjusted. Therefore, it is not adopted in the present invention.

  The method (2) appropriately distributes the code amount to two videos by statistical multiplexing, and will be described as Embodiment 1 of the present invention.

  The method (3) allocates a code amount by using two videos as one stream, and will be described as a second embodiment of the present invention.

(Embodiment 1)
FIG. 1 shows a block diagram of a video encoding apparatus 100 using statistical multiplexing. The video encoding device 100 includes a first video encoding unit 110, a second video encoding unit 120, a multiplexing unit 130, and a GOP control unit 140.

  The high-resolution video that is the first input signal (first input signal) is input to the first video encoding unit 110. The first video encoding unit 110 compresses and encodes the high resolution video of the first input signal. As the compression encoding technique, for example, any encoding technique using inter-frame prediction such as MPEG-2, MPEG-4 AVC, and MPEG-H HEVC standardized by ISO / IEC can be used. The first video encoding unit 110 outputs the encoded video signal to the multiplexing unit 130 and information on the GOP structure of the encoded video signal (reference structure for motion estimation / compensation at the time of encoding). Is output to the GOP control unit 140.

  The low-resolution video that is the second input signal (second input signal) is input to the second video encoding unit 120. When the first input signal and the second input signal are the same video content, a down-conversion unit 150 that converts a high-resolution video into a low-resolution video is provided, the first input signal is down-converted, and the second input signal Can be created.

  The down-conversion unit 150 creates a low-resolution video by reducing the high-resolution video as the first input signal to, for example, a horizontal and vertical resolution of ½. An existing filter such as Lanczos 3 can be used as a filter to be applied at the time of reduction, and the down-conversion method is not particularly limited.

  The second video encoding unit 120 compresses and encodes the low resolution video of the second input signal. As the compression encoding technique, for example, any encoding technique using inter-frame prediction such as MPEG-2, MPEG-4 AVC, and MPEG-H HEVC standardized by ISO / IEC can be used. It is desirable that the encoding technique be a scheme that uses the same GOP processing unit as the encoding unit 110. The second video encoding unit 120 outputs the encoded video signal to the multiplexing unit 130 and information on the GOP structure of the encoded video signal (reference structure for motion estimation / compensation during encoding). Is output to the GOP control unit 140.

  The multiplexing unit 130 multiplexes and transmits two encoded video signals that are the outputs of the first video encoding unit 110 and the second video encoding unit 120. Further, the multiplexing unit 130 observes the encoded information amount generated in the first video encoding unit 110 and the second video encoding unit 120, and when the total instantaneous generated information amount exceeds the threshold, A command is output to the GOP control unit 140. The threshold value is set as a predetermined value that is an upper limit value so that the instantaneous total generated information amount can be suppressed below the bandwidth (bit rate) of the transmission path.

  The GOP control unit 140 receives the GOP structure information of the video encoding from the first video encoding unit 110 and the second video encoding unit 120, and the GOP control unit 140 receives the structure information of the first video encoding unit 110 and the second video encoding unit 120. It manages the mutual relationship of GOP structure. Further, based on a command from the multiplexing unit 130, a control signal is output to at least one of the first video encoding unit 110 and the second video encoding unit 120, and the first video encoding unit 110 and / or the second video encoding unit 120 is output. Control is performed to change the phase of the GOP structure (the reference structure for motion estimation / compensation during encoding) during the encoding process of the video encoding unit 120.

  The operation of the GOP control unit 140 will be described in more detail.

  GOP (Group Of Picture) is a processing unit of video compression / decompression. Generally, only from an intra picture (I) encoded by intra prediction without performing inter-frame prediction and a past frame in the time direction. It consists of three types of inter-prediction (P) for prediction in the forward direction and bi-prediction inter-picture (B) for prediction from a plurality of frames in the time direction. The GOP structure includes, for example, a structure in which I and P are inserted at a certain period, such as IBBPBBPBBIBBP..., And B between them. , One GOP is defined from an intra picture to the next intra picture. A reference relationship (reference structure) for motion estimation / compensation in the GOP is referred to as a GOP structure. The insertion period of an intra picture is generally set periodically, such as 0.5 seconds or 1 second, but control such as forcibly making an intra picture at the time of video cut change may be performed. Intra pictures tend to increase the amount of information generated by encoding as much as inter-frame prediction is not performed.

  Since the high-resolution video and the low-resolution video of the same content are originally the same video, if the GOP structure and the position (phase) of the intra picture are the same, the tendency of the amount of generated information in each picture is almost the same. . FIG. 2 shows the amount of information generated when a high-resolution video and a low-resolution video are encoded by the GOP structure of the same IBBP. In this case, as shown in FIG. 2, the amount of information increases with I, and then information by encoding is generated in the order of P and B. When high-resolution video and low-resolution video are encoded and transmitted at the same time, if the amount of information generated tends to be as shown in FIG. 2, the amount of information increases and decreases at the same time (frame).

  Therefore, the present invention observes the amount of encoding occurrence information of high-resolution video and low-resolution video at the same time (frame) in the multiplexing unit 130 that performs statistical multiplexing, and when the sum of both exceeds a threshold value, A command to change the reference relationship of at least one GOP of the high resolution video and the low resolution video is output to the GOP control unit 140. The GOP control unit 140 outputs a control signal to the first video encoding unit 110 and / or the second video encoding unit 120 to change a reference structure for motion estimation / compensation when encoding video. Specifically, the time (frame) of the reference intra picture is shifted without changing the periodic structure of the reference relationship of motion estimation / compensation (for example, a combination such as IBBP). This means that the phase of the GOP structure is shifted or the phase is changed.

  FIG. 3 shows an example in which the time of an intra picture of a low-resolution video is delayed by two frames from that of a high-resolution video. By shifting the time of intra pictures of two videos without changing the GOP periodic structure, the total amount of generated information at the same time can be reduced and the amount of generated information can be smoothed.

  Note that in an encoding device or a multiplexing device, a buffer may be provided in order to implement the total amount of generated information in a plurality of frames or GOP units. Therefore, it is desirable to determine whether or not to change the GOP structure in the multiplexing unit 130 in which a normal buffer is provided. When the buffer is provided in a plurality of encoding units, the amount of information in the plurality of buffers is determined. You may judge with reference.

  Next, a procedure for determining whether or not to change the phase of the GOP structure and how many frames to shift will be described.

  FIG. 4 shows a flowchart for controlling the phase of the GOP structure of the encoding unit.

  First, in step 1 (S1), the first video encoding unit 110 and the second video encoding unit 120 compress and encode the high resolution video and the low resolution video, respectively. In the first stage, video encoding is performed with continuity with the GOP structure used so far in each video encoding unit.

  In step 2 (S2), the amounts of information generated at the same time (frame) of the first video encoding unit 110 and the second video encoding unit 120 are summed, and the sum of both is compared with a threshold value. If the total amount of generated information exceeds the threshold value, the process proceeds to step 3 (S3). If the total amount of generated information does not exceed the threshold value, the process is temporarily terminated, and S1 and S2 are repeated to continuously monitor the total amount of generated information.

  In step 3 (S3), the amount of change in the phase of the GOP structure is set. That is, a GOP phase control signal for shifting the intra picture (I) by a predetermined number of frames (shift amount per time) is output. As the video encoding device, when the total amount of generated information exceeds a threshold, the multiplexing unit 130 outputs a command to change the phase of the GOP structure of the video encoding unit, and the GOP control unit 140 Based on this command, a control signal for changing the phase of the GOP structure (reference structure for motion estimation / compensation) to a predetermined number of frames is output to the first video encoding unit 110 and / or the second video encoding unit 120. For example, the GOP control unit 140 outputs a control signal for shifting the position of the GOP intra-picture to the second video coding unit 120 by a predetermined number of frames (for example, one frame). This method of shifting the position (phase) may be delayed or advanced with respect to one, and may be shifted by any number of frames per time.

  Next, in step 4 (S4), in the video encoding unit (for example, the second video encoding unit 120), the GOP phase of the encoding process, that is, the position (time) of the intra picture (I) is set in S3. Control to change to the set frame. And it returns to S1 again.

  In S1, video encoding is performed according to a newly set GOP structure (motion estimation / compensation reference structure) for the video frame. In S2, the first video encoding unit 110 and the second video encoding based on the new settings are performed. The sum total of the amount of information generated by the encoding unit 120 is compared with a threshold value.

  If the total amount of generated information still exceeds the threshold for this comparison result, the process of S3 → S4 → S1 → S2 is repeated. Each time the process is repeated, the position of the intra picture at the time of video coding is shifted by a predetermined number of frames.

  If the maximum value of the total amount of generated information for each frame at that time is equal to or less than a predetermined threshold for the comparison result of S2, it is decided to shift the phase of the GOP structure by that frame, and the process ends.

  Finally, the signal encoded after changing the phase of the GOP structure is multiplexed by the multiplexing unit 130 and output. Further, the monitoring of the total amount of generated information in S2 is continued thereafter.

  In the flowchart of FIG. 4, the determination condition of S3 is when the generated information amount exceeds the threshold, but it may be when the generated information amount exceeds the threshold, and the generated information amount exceeds the bandwidth of the transmission path. The condition for detecting this may be set as appropriate.

  In this embodiment, the input signal is the same video (content) of two different resolutions, but may be three or more different resolution videos. At this time, the video encoding device is provided with a video encoding unit corresponding to the number of inputs. In addition, when the total amount of encoding generation information exceeds a threshold, the phase of an arbitrary GOP structure among a plurality of video encoding units can be adjusted. At this time, it is possible to set a phase shift method for each video encoding unit. For example, the second video encoding unit uses the position of the intra picture (I) of the first video encoding unit as a reference. The position is shifted by a first predetermined number of frames (for example, one frame), and the third video encoding unit shifts the position of the intra picture by a second predetermined number of frames (for example, two frames) to code It is possible to adjust the sum of the amount of generated information.

  Further, in this embodiment, even when two types of images with different resolutions are images of different contents, the operation and effect of enabling transmission within the band can be obtained. Furthermore, even when the resolutions of the first input signal and the second input signal are the same, the video encoding device is similarly effective.

(Embodiment 2)
The second embodiment of the present invention will be described below. The video encoding apparatus according to the second embodiment of the present invention uses hierarchical encoding.

  FIG. 5 shows a block diagram of a video encoding apparatus 200 using hierarchical encoding. The video encoding device 200 includes a first video encoding unit 110, a second video encoding unit 120, a multiplexing unit 130, a GOP control unit 140, a base layer video local decoding unit 160, and an up-conversion unit 170. With.

  The high-resolution video that is the first input signal (first input signal) is input to the first video encoding unit 110. The first video encoding unit 110 compresses and encodes the high-resolution video of the first input signal. Here, the high-resolution video is the first video encoding unit as an enhancement layer (EL) of hierarchical encoding. 110, the first video encoding unit 110 performs an enhancement layer encoding process. At this time, the base layer video (the video obtained by decoding the base layer encoded signal and expanding it by the up-conversion unit 170) can be used as one of the prediction candidates for the extended layer encoding. As the compression encoding technique, for example, any encoding technique using inter-frame prediction such as MPEG-2, MPEG-4 AVC, and MPEG-H HEVC standardized by ISO / IEC can be used. The first video encoding unit 110 outputs the encoded video signal (enhancement layer encoded signal) to the multiplexing unit 130 and also the GOP structure of the encoded video signal (motion estimation / encoding at the time of encoding). The information regarding the compensation reference structure) is output to the GOP control unit 140.

  The low-resolution video that is the second input signal (second input signal) is input to the second video encoding unit 120. When the first input signal and the second input signal are the same video content, a down-conversion unit 150 that converts a high-resolution video into a low-resolution video is provided, the first input signal is down-converted, and the second input signal Can be created.

  The down-conversion unit 150 creates a low-resolution video by reducing the high-resolution video as the first input signal to, for example, a horizontal and vertical resolution of ½. An existing filter such as Lanczos 3 can be used as a filter to be applied at the time of reduction, and the down-conversion method is not particularly limited.

  The second video encoding unit 120 compresses and encodes the low resolution video of the second input signal. Here, the low resolution video is used as a base layer (BL: Base Layer) for hierarchical encoding. The second video encoding unit 120 performs base layer encoding processing. As the compression encoding technique, for example, any encoding technique using inter-frame prediction such as MPEG-2, MPEG-4 AVC, and MPEG-H HEVC standardized by ISO / IEC can be used. It is desirable that the encoding technique is a scheme using the same GOP configuration as that of the encoding unit 110. The second video encoding unit 120 outputs the encoded video signal (base layer encoded signal) to the multiplexing unit 130 and the base layer video local decoding unit 160, and the structure of the GOP of the encoded video signal. Information related to (reference structure for motion estimation / compensation during encoding) is output to the GOP control unit 140.

  The base layer video local decoding unit 160 decodes the base layer encoded signal which is the output of the second video encoding unit 120, and generates a low-resolution video (base layer) of the second input signal. However, since the generated video is decoded after the information is compressed by the encoding, it is affected by the GOP structure at the time of encoding (whether it is compressed by I, P, or B). It does not completely restore the low resolution video of the input signal.

  The up-conversion unit 170 up-converts the low-resolution video (base layer) obtained by decoding by the base layer video local decoding unit 160, expands it to the original resolution, and outputs the second encoding unit (enhancement layer coding). Part) 110. Existing filters such as Lanczos3 can be used for the filter used for enlargement, and any filter can be used. Note that the enlarged video is not the same as the original video because it has been enlarged after the original video has been reduced and encoded and decoded.

  The operations of multiplexing section 130 and GOP control section 140 are basically the same as in the first embodiment. That is, the multiplexing unit 130 multiplexes and transmits two encoded video signals that are the outputs of the first video encoding unit 110 and the second video encoding unit 120. In addition, the multiplexing unit 130 encodes information generated by the first video encoding unit 110 and the second video encoding unit 120 (encoding generation information of the base layer video and the extended layer video at the same time (frame)). When the total instantaneous generated information amount exceeds the threshold value, a command is output to the GOP control unit 140.

  The GOP control unit 140 receives the GOP structure information of the video encoding from the first video encoding unit 110 and the second video encoding unit 120, and the GOP control unit 140 receives the structure information of the first video encoding unit 110 and the second video encoding unit 120. It manages the mutual relationship of GOP structure. Further, based on a command from the multiplexing unit 130, a control signal is output to at least one of the first video encoding unit 110 and the second video encoding unit 120, and the first video encoding unit 110 and / or the second video encoding unit 120 is output. Control is performed to change the phase of the GOP structure (the reference structure for motion estimation / compensation during encoding) during the encoding process of the video encoding unit 120 by a predetermined number of frames.

  Hierarchical encoding will be briefly described with reference to FIG. Hierarchical coding can use a standardized method as an extension of the coding technology such as MPEG-2, MPEG-4 AVC, or MPEG-H HEVC. For example, in MPEG-H HEVC, a video decoded from a base layer encoded signal and expanded by the up-conversion unit 170 is used as one of prediction candidates in the case of enhancement layer coding. The arrows in FIG. 6 indicate the reference relationship for motion estimation / compensation. In the case of hierarchical coding, the base layer (BL) is processed in the same way as normal coding, but the enhancement layer (EL) is composed of prediction from BL and prediction from other EL frames. .

  FIG. 7 shows an example in which the time of the intra-picture of the extended layer video is delayed by one frame from the basic layer video. By shifting the time of the intra pictures of the two videos, it is possible to reduce the total amount of generated information at the same time. Note that the determination of whether to change the phase of the GOP structure and the determination of how many frames to shift can be similarly performed according to the flowchart of FIG. The comparison between the amount of generated information and the threshold value is desirably performed by the multiplexing unit 130 in which a normal buffer is provided, but in the first video encoding unit (enhancement layer encoding unit) in which buffers for both encoded outputs are provided. It is also possible to implement.

  Further, as indicated by the thick line in FIG. 7, there is a frame that uses a BL I picture or P picture for prediction of an EL B picture. When the GOP structure is the same in BL and EL as in the conventional hierarchical encoding process of FIG. 6, EL B is predicted from BL B. In the present invention (FIG. 7), I and I By performing the prediction from P, it is possible to use a reference frame having a higher image quality than the prediction from B, and as a result, there is an effect of improving the image quality of EL B.

  A signal encoded after changing the phase of the GOP structure is multiplexed by the multiplexing unit 130 and output.

  In this embodiment, the input signal is the same video (contents) of two types of different resolutions of the basic layer video and the extended layer video, but further, the intermediate layer video is added and three or more types of different resolutions are added. Video may be targeted. At this time, the video encoding device is provided with a video encoding unit corresponding to the number of inputs, and the lower video encoding unit includes a local decoding unit that decodes the video encoded signal, and the decoded video. An up-conversion unit that performs up-conversion and outputs the result to a higher-level video encoding unit is provided. In addition, when the total amount of encoding generation information exceeds a threshold, the phase of an arbitrary GOP structure among a plurality of video encoding units can be adjusted. At this time, it is possible to set a phase shift method for each video encoding unit, and to adjust the total amount of encoding generated information.

  In the second embodiment, when the HEVC hierarchical encoding process is used, even if two types of videos with different resolutions are videos of different contents, they can function similarly as a video encoding device. it can.

  Note that a computer can be suitably used to function as the above-described video encoding apparatus, and such a computer stores a program describing processing contents for realizing each function of the video encoding apparatus. This program can be realized by reading out and executing this program by a central processing unit (CPU) of the computer. This program can be recorded on a computer-readable recording medium.

  Although the above embodiment has been described as a representative example, it will be apparent to those skilled in the art that many changes and substitutions can be made within the spirit and scope of the invention. Therefore, the present invention should not be construed as being limited by the above-described embodiments, and various modifications and changes can be made without departing from the scope of the claims. For example, a plurality of constituent blocks and steps described in the embodiments can be combined into one or divided.

100 video encoding device 110 first video encoding unit 120 second video encoding unit 130 multiplexing unit 140 GOP control unit 150 down-conversion unit 160 base layer video local decoding unit 170 up-conversion unit 200 video encoding device

Claims (7)

In a video encoding device for simultaneously compressing and multiplexing a plurality of different videos,
A code that is generated at the same time by changing the phase of the motion estimation / compensation reference structure at the time of encoding at least one of the plurality of different videos when the amount of coded information to be generated exceeds a predetermined value A video encoding apparatus characterized by smoothing the amount of encoded information below the predetermined value. The video encoding device according to claim 1, wherein
A plurality of video encoding units;
A multiplexer that multiplexes outputs of the plurality of video encoding units and outputs a command when the amount of encoded information generated in the plurality of video encoding units exceeds the predetermined value;
A GOP control unit that changes a phase of a reference structure for motion estimation / compensation at the time of encoding of at least one video encoding unit of the plurality of video encoding units based on the command;
A video encoding device comprising: The video encoding device according to claim 1 or 2,
The plurality of video encoding units includes a first video encoding unit that encodes a high-resolution video, and a second video encoding unit that encodes a low-resolution video,
The video encoding device, wherein the input video signal to the second video encoding unit is a video signal obtained by down-converting the input video signal to the first video encoding unit. A video encoding device that simultaneously compresses and multiplexes a plurality of videos having different resolutions, and the compression encoding is a compression encoding having a structure for predicting a high-resolution video from a basic video. In a video encoding device,
When the amount of generated encoded information exceeds a predetermined value, the phase of the motion estimation / compensation reference structure is changed at the time of encoding at least one of the plurality of different resolution images, and is generated at the same time. A video encoding apparatus characterized by smoothing an encoded information amount to be equal to or less than the predetermined value. The video encoding device according to claim 4, wherein
A plurality of video encoding units including a video encoding unit for encoding a basic video and a video encoding unit for encoding a high-resolution video;
Decoding the output of the video encoding unit that encodes the basic video, and generating a low-resolution video; and
An up-conversion unit for enlarging the low-resolution video and outputting the high-resolution video to a video encoding unit;
A multiplexer that multiplexes outputs of the plurality of video encoding units and outputs a command when the amount of encoded information generated in the plurality of video encoding units exceeds the predetermined value;
A GOP control unit that changes a phase of a reference structure for motion estimation / compensation at the time of encoding of at least one video encoding unit of the plurality of video encoding units based on the command;
A video encoding device comprising: In the video coding device according to any one of claims 1 to 5,
When the generated encoded information amount exceeds a predetermined value, the phase of the motion estimation / compensation reference structure at the time of encoding at least one of the plurality of video encoding units is shifted by a predetermined number of frames. It is determined whether the amount of encoded information is within the predetermined value, and if it is within the predetermined value, the reference structure is set to the shifted phase, and if it exceeds the predetermined value, the predetermined number of frames is further shifted. A video encoding device.   A program for causing a computer to function as the video encoding device according to any one of claims 1 to 6.

Images