JP6574148B2 - Transcoding system, transcoding method, computer-readable recording medium, and encoding apparatus - Google Patents

Description

  The present invention relates to a transcoding system, a transcoding method, a computer-readable recording medium, a decoding device, and an encoding device.

  A moving picture coding scheme using intra prediction (intraframe prediction) or inter prediction (interframe prediction), residual conversion, and entropy coding has been proposed. Generating moving image bitstreams having different settings from moving image bitstreams generated by such moving image encoding is called transcoding. For example, transcoding is used when reducing only the bit rate while maintaining the same resolution, or when generating a video with reduced resolution.

  Non-Patent Document 1 discloses a method for suppressing quantization noise by limiting quantization parameters used for re-quantization with respect to transcoding conforming to MPEG-2.

“Basic study on MPEG-2 video transcoder re-quantization code amount control method considering distortion increase characteristics due to reduced code amount”, Isao Nagayoshi et al., Journal of the Institute of Image Information and Television Engineers, Vol. 55 (2001) No. 12, 2001

  However, in the method of using MPEG-2 as an input bit stream as described in Non-Patent Document 1, smoothing by a deblocking filter is not considered. Therefore, there is a concern about performance degradation due to too strong restrictions in requantization.

  The present invention has been made in view of these problems, and an object of the present invention is to provide a technique that can achieve both reduction in the amount of processing in transcoding and suppression of image quality deterioration due to transcoding.

One embodiment of the present invention relates to a transcoding system. The transcoding system includes a first acquisition unit that acquires a bit stream of a moving image, an output unit that outputs a new bit stream having a setting different from that of the acquired bit stream, and entropy-decodes the acquired bit stream. Based on the result of the entropy decoding, the decoding means, the second acquisition means for acquiring the motion vector, the amount of the error signal, and the compression mode of the motion vector for each block, based on the acquired amount of the error signal, a new When generating a bitstream, a first determination unit that determines whether or not to apply the acquired motion vector to the corresponding block to be encoded, and a new bit based on the acquired compression mode When generating a stream, a second judgment is made to determine whether or not to apply the acquired compression mode to the encoding target block. And means, the.

  It should be noted that any combination of the above-described constituent elements, or those obtained by replacing the constituent elements and expressions of the present invention with each other between apparatuses, methods, systems, computer programs, recording media storing computer programs, and the like are also included in the present invention. It is effective as an embodiment of

  According to the present invention, it is possible to reduce both the amount of processing in transcoding and the suppression of image quality deterioration due to transcoding.

It is a schematic diagram which shows the structure of a cable television system provided with the transcoding apparatus which concerns on embodiment. It is a block diagram which shows the function and structure of the transcoding apparatus of FIG. It is a block diagram which shows the function and structure of the 1st inter estimation part of FIG. It is a block diagram which shows the function and structure of the 2nd inter estimation part of FIG. It is a flowchart which shows a series of processes in the encoding control part of FIG. It is a schematic diagram which shows the structure of the transcoding system which concerns on a modification.

  Hereinafter, the same or equivalent components, members, and processes shown in the drawings are denoted by the same reference numerals, and repeated description is appropriately omitted. In addition, in the drawings, some of the members that are not important for explanation are omitted.

  The transcoding device according to the embodiment extracts a motion vector (Motion Vector, MV), the presence / absence of an error (residual) signal, and a decoded quantization value of a coding unit from an input bitstream decoding result. . The transcoding device sends the same motion vector as the extracted motion vector to the encoding unit and does not send an error signal for a block having no error signal. Further, the transcoding device derives a correction value QPoffset from the difference between the input resolution and the output resolution. The transcoding device sets the coded quantization value applied to the coding unit to be equal to or greater than the value obtained by adding QPoffset to the decoded quantization value. As a result, it is possible to reduce the amount of processing in the transcoding device by reusing motion vectors while reducing or eliminating the encoding noise itself included in the input bitstream.

  FIG. 1 is a schematic diagram illustrating a configuration of a cable television system 10 including a transcoding device 100 according to an embodiment. The cable television system 10 includes a broadcaster server 12, a center facility 14, a cable network 16, and an end user 18. The broadcaster server 12 is operated by a broadcaster and provides a relatively high resolution or high bit rate moving image to the center facility 14 by wire or wirelessly. The center facility 14 is connected to the broadcaster server 12 and the cable network 16. The center facility 14 includes the transcoding device 100 according to the embodiment. The center facility 14 acquires a moving image from the broadcast provider server 12, and changes the settings of the acquired moving image, that is, the resolution and the bit rate so as to meet the specifications of the cable network 16 using the transcoding device 100. The center facility 14 transmits the moving image obtained as a result of transcoding to the cable network 16. Each of the plurality of end users 18 is connected to the cable network 16 and reproduces a moving image acquired from the cable network 16 using a set-top box or another decoding device. For example, the resolution of the moving image transmitted by the broadcast provider server 12 is 4K and the bit rate is 35 Mbps, the resolution of the moving image transmitted from the center facility 14 to the cable network 16 is HD (High Definition), and the bit rate is 10 Mbps. It is.

  The cable television system 10 is an example, and the transcoding device 100 according to the embodiment can be applied to any system or service including decoding of a moving image and subsequent encoding. In such a system or service, the Internet, a telephone network, or the like may be used instead of the cable network 16.

  More specifically, the broadcaster server 12 encodes a moving image to generate an input bit stream BSI. The transcoding device 100 acquires the input bit stream BSI from the broadcaster server 12, decodes the acquired input bit stream BSI, and encodes it with different settings to generate an output bit stream BST. The transcoding device 100 outputs the generated output bit stream BST to the cable network 16. The end user 18 obtains the output bit stream BST from the cable network 16 and decodes it. Both the input bit stream BSI and the output bit stream BST have a structure compliant with HEVC (High Efficiency Video Coding).

  FIG. 2 is a block diagram showing the function and configuration of the transcoding device 100 of FIG. Each block shown in FIG. 2 and the subsequent block diagrams can be realized in hardware by an element such as a CPU of a computer or a mechanical device, and can be realized by a computer program or the like in software. , Depicts functional blocks realized by their cooperation. Therefore, those skilled in the art who have touched this specification will understand that these functional blocks can be realized in various forms by a combination of hardware and software.

  The transcoding device 100 includes a decoding unit 102, an encoding unit 104, a motion processing unit 106, and a resolution conversion unit 146. The decoding unit 102 acquires the input bitstream BSI, decodes the acquired input bitstream BSI, reconstructs a moving image, and outputs the reconstructed moving image to the resolution conversion unit 146. The resolution conversion unit 146 converts the resolution of the reconstructed moving image using a known resolution conversion technique. The resolution conversion unit 146 outputs the reconstructed moving image whose resolution has been converted to the encoding unit 104. The encoding unit 104 acquires a reconstructed moving image, encodes the acquired reconstructed moving image to generate an output bitstream BST that is a new bitstream, and connects the generated output bitstream BST to a cable. Output to the network 16.

  The motion processing unit 106 acquires information on the reuse of motion vectors from the result of the parsing processing of the input bitstream BSI in the decoding unit 102. Based on the acquired information, the motion processing unit 106 acts on the encoding unit 104 to reuse the motion vector and skip the motion search when the image quality is ensured even when the motion vector is reused. To do. The motion processing unit 106 also performs the coded quantization value of the quantization in the coding unit 104 based on the decoded quantization value obtained from the result of the parsing process described later and the resolution ratio between input and output in the resolution conversion unit 146. To control.

  The decoding unit 102 includes an entropy decoding unit 108, a first frame buffer 110, a first in-loop filter 112, a first inverse quantization / inverse transform unit 114, a first inter prediction unit 116, and a first intra prediction. Part 118 and a first addition part 120.

  The entropy decoding unit 108 receives the input bit stream BSI, entropy decodes the received input bit stream BSI, and extracts a level value and side information. Note that the process of obtaining side information and level values from the bitstream is referred to as a parse process. Reconstructing the pixel value using the side information and the level value obtained in this way is called a decoding process.

  A moving image frame represented by the input bitstream BSI is divided into a plurality of blocks. The block sizes vary, and the plurality of blocks have a quadtree structure. The division of the frame into blocks in the present embodiment is in accordance with the block division in HEVC. That is, the size of CTU (Coding Tree Unit), which is the largest processing size, is 64 × 64, and there are 32 × 32, 16 × 16, and 8 × 8 sizes obtained by repeatedly dividing this into four.

  Side information is information necessary for reconstruction of pixel values used in a decoding device provided in the decoding unit 102 or the end user 18, and a prediction mode indicating whether intra prediction or inter prediction is used, It includes related information such as motion information, quantized values, and block size.

  In order to reduce the amount of codes during transmission, motion vectors used in inter prediction are basically compressed and transmitted. There are at least two motion vector compression modes, a merge mode and an advanced prediction mode.

In merge mode, processing is performed by selecting motion vectors of several encoded blocks around the block to be processed as candidates, selecting one motion vector as a merge destination from the candidates, and reusing the selected motion vector. A motion vector of the target block is generated. In this case, the motion information includes a set of a merge identifier and a reference frame identifier that are identifiers of motion vector candidates selected as a merge destination in the inter prediction of encoding.
In the advanced prediction mode, motion vectors of several encoded blocks around the processing target block are used as candidates, a prediction motion vector is selected from the candidates, and a difference motion vector is added to the selected prediction motion vector. Then, a motion vector of the processing target block is generated. In this case, the motion information includes a set of a prediction identifier, a difference motion vector, and a reference frame identifier that are identifiers of motion vector candidates selected as prediction motion vectors in the inter prediction of encoding.

  The first inverse quantization / inverse transform unit 114 generates an error signal by performing inverse quantization and inverse transform on the level value obtained as a result of the parsing process in the entropy decoding unit 108 for each block. The first addition unit 120 determines whether either the first inter prediction unit 116 or the first intra prediction unit 118 depends on whether the block to be decoded is intra-predicted or inter-predicted. The prediction image of the decoding target block to be output and the error signal of the decoding target block generated by the first inverse quantization / inverse transform unit 114 are added to reproduce the decoding target block. The first addition unit 120 outputs the reproduced block to be decoded to the first intra prediction unit 118 and the first in-loop filter 112.

  The first inter prediction unit 116 receives the reference frame of the block to be decoded from the first frame buffer 110. The first inter prediction unit 116 generates and outputs a predicted image of the decoding target block by inter-frame prediction based on the reference frame. The first intra prediction unit 118 performs intra-frame prediction using the already reproduced block of the frame to which the decoding target block belongs, and generates and outputs a predicted image of the decoding target block. The first in-loop filter 112 is a deblock filter, for example. The first in-loop filter 112 generates a locally decoded image corresponding to the frame and outputs it to the first frame buffer 110. This locally decoded image is used for inter-frame prediction in the first inter prediction unit 116, and at the same time, is output to the resolution conversion unit 146 as a frame of a reconstructed moving image.

  FIG. 3 is a block diagram illustrating the function and configuration of the first inter prediction unit 116. The first inter prediction unit 116 includes an MV restoration unit 302, an MV buffer 304, and a predicted image generation unit 306. The MV restoration unit 302 acquires the motion information of the block to be decoded from the entropy decoding unit 108, and restores the motion vector of the block to be decoded based on the acquired motion information. The MV restoration unit 302 registers the restored motion vector and the reference frame identifier included in the acquired motion information in association with the block to be decoded in the MV buffer 304 and passes it to the predicted image generation unit 306.

In the merge mode, the MV restoration unit 302 refers to the MV buffer 304 and specifies a motion vector candidate specified by the merge identifier included in the motion information. The MV restoration unit 302 generates a motion vector of a decoding target block by reusing the identified motion vector candidates. The MV restoration unit 302 outputs the generated motion vector and the reference frame identifier included in the motion information to the MV buffer 304 and the predicted image generation unit 306.
In the case of the advanced prediction mode, the MV restoration unit 302 refers to the MV buffer 304 and specifies a motion vector candidate specified by the prediction identifier included in the motion information. The MV restoration unit 302 generates a motion vector of the decoding target block by adding the identified motion vector candidate and the difference motion vector included in the motion information. The MV restoration unit 302 outputs the generated motion vector and the reference frame identifier included in the motion information to the MV buffer 304 and the predicted image generation unit 306.

  The predicted image generation unit 306 acquires the reference frame specified by the reference frame identifier acquired from the MV restoration unit 302 from the first frame buffer 110. The predicted image generation unit 306 generates a predicted image of the decoding target block from the motion vector acquired from the MV restoration unit 302 and the reference frame acquired from the first frame buffer 110.

  Returning to FIG. 2, the encoding unit 104 includes a second frame buffer 122, a second in-loop filter 124, a second inter prediction unit 126, a second intra prediction unit 128, a subtraction unit 130, and a second addition. Unit 132, transform / quantization unit 134, second inverse transform / inverse quantization unit 136, and entropy coding unit 138. The frame of the reconstructed moving image after resolution conversion output from the resolution conversion unit 146 is input to the subtraction unit 130, the second inter prediction unit 126, and the second intra prediction unit 128 for each block.

  The second inter prediction unit 126 receives the reference frame of the block to be encoded from the second frame buffer 122. The second inter prediction unit 126 generates and outputs a prediction image of a block to be encoded by inter-frame prediction based on the reference frame. The second intra prediction unit 128 receives the image data of the already processed block of the encoding target frame from the second addition unit 132. The second intra prediction unit 128 generates and outputs a prediction image of the encoding target block based on the other blocks of the encoding target frame. A prediction image output from the second inter prediction unit 126 or a prediction image output from the second intra prediction unit 128 depending on whether inter-frame prediction or intra-frame prediction is applied to the block to be encoded Is output to the subtractor 130 and the second adder 132.

  The subtraction unit 130 generates an error signal between the encoding target image and the predicted image in units of blocks. The subtraction unit 130 outputs an error signal indicating an error between the block to be encoded and the prediction image output by the second inter prediction unit 126 or the prediction image output by the second intra prediction unit 128. The transform / quantization unit 134 transforms (for example, orthogonal transform) and quantizes the error signal in units of blocks, and generates a level value. The transform / quantization unit 134 outputs the generated level value to the entropy coding unit 138 and the second inverse transform / inverse quantization unit 136. The entropy encoding unit 138 entropy encodes the level value generated by the transform / quantization unit 134 and side information such as motion information provided from the second inter prediction unit 126 to generate an output bit stream BST. ,Output.

  The second inverse transform / inverse quantization unit 136 performs processing reverse to the processing in the transform / quantization unit 134 to generate (regenerate) an error signal. The second addition unit 132 receives the error signal output from the second inverse transform / inverse quantization unit 136 and the prediction image output from the second inter prediction unit 126 or the prediction image output from the second intra prediction unit 128. The blocks to be encoded are generated by addition and output to the second intra prediction unit 128 and the second in-loop filter 124. The second in-loop filter 124 generates a locally decoded image corresponding to the encoding target frame and outputs it to the second frame buffer 122. This locally decoded image is used for inter-frame prediction in the second inter prediction unit 126.

  The motion processing unit 106 includes an acquisition unit 140, a determination unit 142, and an encoding control unit 144. The acquisition unit 140 acquires, for each block, motion information, a motion vector, a motion vector compression mode, the presence / absence of an error signal, and a decoded quantization value from the result of entropy decoding of the input bitstream BSI in the entropy decoding unit 108. . In particular, the decoded quantization value is acquired in units of CU (Coding Unit). Basically, motion vectors are not directly included in entropy decoding results. The acquisition unit 140 may generate a motion vector by a process similar to the restoration process in the MV restoration unit 302 of the first inter prediction unit 116 described with reference to FIG. Alternatively, the acquisition unit 140 may acquire a motion vector with reference to the MV buffer 304 of the first inter prediction unit 116.

  Based on the presence / absence of an error signal, the determination unit 142 determines whether to apply the motion vector acquired by the determination unit 142 to the corresponding encoding target block when generating the output bitstream BST. To do. The determination unit 142 determines to apply if there is no error signal, and determines not to apply otherwise. If the determination unit 142 determines to apply, the motion information, the motion vector, and the compression mode acquired by the acquisition unit 140 are output to the encoding control unit 144. If the determination unit 142 determines not to apply, the determination unit 142 outputs an invalid value to the encoding control unit 144. Hereinafter, the output of the determination unit 142 is referred to as a determination output. If there is an error signal, an invalid value is output because an appropriate motion vector may be found by a new motion vector search by re-encoding. When there is no error signal, it can be considered that the prediction is accurately performed by motion compensation prediction. Therefore, it is better to use the same motion vector in re-encoding and not send a prediction error signal.

  The encoding control unit 144 controls inter prediction of the block to be encoded in the encoding unit 104 based on the determination output of the determination unit 142. Hereinafter, the function of the encoding control unit 144 will be described together with the configuration of the second inter prediction unit 126.

  FIG. 4 is a block diagram illustrating the function and configuration of the second inter prediction unit 126. The second inter prediction unit 126 includes a motion information generation unit 402, an MV buffer 404, a predicted image generation unit 406, and a motion search unit 408. The motion search unit 408 acquires an encoding target block. The motion search unit 408 refers to the second frame buffer 122 and determines a motion vector and a reference frame of a block to be encoded using a known motion estimation technique. At the time of this determination, an optimal motion vector compression mode is selected according to the coding cost or the like. The motion search unit 408 passes the determined motion vector and the reference frame identifier that specifies the determined reference frame to the MV buffer 404 and the predicted image generation unit 406.

The MV buffer 404 holds the motion vector and reference frame identifier acquired from the motion search unit 408 in association with the block to be encoded.
The predicted image generation unit 406 acquires a reference frame specified by the reference frame identifier acquired from the motion search unit 408 from the second frame buffer 122. The predicted image generation unit 406 generates a predicted image of the block to be encoded from the motion vector acquired from the motion search unit 408 and the reference frame acquired from the second frame buffer 122.

  The motion information generation unit 402 identifies which compression mode has been selected by the motion search unit 408. When the identified compression mode is the merge mode, the motion information generation unit 402 generates motion information including the merge identifier and the reference frame identifier determined by the motion search unit 408 and outputs the motion information to the entropy encoding unit 138. To do. When the identified compression mode is the advanced prediction mode, the motion information generation unit 402 generates motion information including the prediction identifier, the difference motion vector, and the reference frame identifier determined by the motion search unit 408, and generates an entropy code. To the conversion unit 138.

  The above is the processing of the second inter prediction unit 126 when it is determined that the determination unit 142 does not apply. That is, when the encoding control unit 144 receives an invalid value as a determination output, the encoding control unit 144 does not act on the second inter prediction unit 126. When it determines with applying in the determination part 142, the 2nd inter prediction part 126 performs inter prediction with respect to the block of encoding using the motion vector acquired by the acquisition part 140. FIG. That is, when the encoding control unit 144 receives the motion information, the motion vector, and the compression mode as the determination output, the encoding control unit 144 performs the second inter prediction unit so as to reuse the motion vector without performing the motion search. 126 is controlled.

  More specifically, the encoding control unit 144 deactivates the motion search unit 408 so that the motion search is not performed on the block to be encoded when it is determined by the determination unit 142 to apply. In addition, the encoding control unit 144 passes the motion vector and the reference frame identifier included in the determination output to the MV buffer 404 and the predicted image generation unit 406, and the compression mode and the motion information included in the determination output are transferred to the motion information generation unit. 402. As a result, even when the motion search unit 408 is deactivated, the motion information generation unit 402, the MV buffer 404, and the predicted image generation unit 406 perform encoding control on information necessary for generating a predicted image and generating motion information. It can be obtained from the part 144.

  Returning to FIG. 2, the encoding control unit 144 further controls the transform / quantization unit 134. The resolution conversion unit 146 is a ratio r_in / r_out (reduction ratio r_out /) of the resolution r_in of the reconstructed moving image input to the resolution conversion unit 146 and the resolution r_out of the reconstructed moving image output from the resolution conversion unit 146. The correction value QPoffset (negative value) proportional to the reciprocal of r_in is output to the encoding control unit 144. The encoding control unit 144 includes (1) a quantized value derived from rate control, and (2) a decoded quantized value + resolution acquired by the acquiring unit 140. The correction value QPoffset acquired from the conversion unit 146 is set to the larger one. In other words, the encoding control unit 144 sets the lower limit of the encoded quantized value based on the decoded quantized value acquired by the acquiring unit 140 and the correction value QPoffset acquired from the resolution converting unit 146. Thereby, encoding of quantization noise of the input bitstream BSI is suppressed.

  For example, when the correction value QPoffset is reduced to ½, the resolution conversion unit 146 outputs -6 as the correction value QPoffset, and when it is reduced to ¼, outputs -12 as the correction value QPoffset. This is because the quantization noise of the signal is reduced by the reduction, and the quantization noise included in the input bit stream BSI is reduced. The quantization value is designed to double the quantization step every time it increases by 6.

  When the resolution conversion unit 146 is not provided, the encoding control unit 144 uses (1) a quantized value derived from rate control as the encoded quantization value in the conversion / quantization unit 134, and (2) an acquisition unit. Set to the larger one of the decoded quantized values acquired by 140.

The operation of the transcoding device 100 configured as above will be described.
FIG. 5 is a flowchart showing a series of processes in the encoding control unit 144 of FIG. The encoding control unit 144 determines whether the determination output of the determination unit 142 is an invalid value for the block to be encoded (S502). If it is an invalid value (YES in S502), the encoding control unit 144 ends the process. If it is not an invalid value (NO in S502), the encoding control unit 144 is acquired by the acquisition unit 140 for the block to be encoded when generating the output bitstream BST based on the compression mode included in the determination output. It is determined whether to apply the compressed mode (S504). More specifically, the encoding control unit 144 determines to apply when the compression mode included in the determination output is the merge mode, and determines not to apply otherwise. When the compression mode included in the determination output is the merge mode (YES in S504), the encoding control unit 144 fixes the motion vector compression mode in the second inter prediction unit 126 to the merge mode (S506). When not in the merge mode (NO in S504) or after the processing in step S506, the encoding control unit 144 deactivates the motion search in the second inter prediction unit 126 (S508). The encoding control unit 144 provides the motion vector and the reference frame identifier included in the determination output to the MV buffer 404 and the predicted image generation unit 406 of the second inter prediction unit 126 instead of the motion search (S510).

  Based on the description in this specification, each unit can be realized by a CPU (not shown), a module of an installed application program, a module of a system program, a semiconductor memory that temporarily stores the contents of data read from a hard disk, or the like. Will be understood by those of ordinary skill in the art having touched this specification.

  According to transcoding apparatus 100 according to the present embodiment, it is possible to suppress the generation of a prediction error signal due to coding noise and the generation of a difference signal due to a different quantization step. As a result, even in the transcoding device 100 that cannot refer to the original image in re-encoding, the encoding of the encoding noise itself included in the input bitstream BSI can be reduced.

  Also, in the transcoding device 100 according to the present embodiment, motion search in re-encoding is deactivated for a block that is often subjected to inter prediction, and a motion vector obtained from the input bitstream BSI is reused. Is done. For a block that is not subjected to much inter prediction, a motion vector is not reused, and a new motion vector is detected in re-encoding. Therefore, it is possible to reduce the amount of processing related to motion search while keeping the image quality of the moving image represented by the output bitstream BST high.

  Also, in transcoding apparatus 100 according to the present embodiment, when merge mode is selected for a decoding target block of input bitstream BSI, merge mode is also selected for a corresponding encoding target block. Is done. In general, the merge mode is often selected for blocks belonging to the same object on the image. Therefore, by fixing the merge mode on the decoding side in the re-encoding, the information “same object” can be taken over, and the image quality can be maintained and improved. Further, since the merge mode has a relatively high code amount reduction effect among the compression modes, the code amount of the output bitstream BST can be reduced by fixing the merge mode.

  The configuration and operation of transcoding apparatus 100 according to the embodiment have been described above. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to each component and combination of processes, and such modifications are within the scope of the present invention.

  In the embodiment, the determination unit 142 uses the presence / absence of an error signal as a determination criterion. For example, the acquisition unit 140 may acquire the amount of the error signal, and the determination unit 142 may compare the acquired amount with a threshold value.

  In the embodiment, the case where the transcoding device 100 includes the decoding unit 102, the encoding unit 104, the motion processing unit 106, and the resolution conversion unit 146 has been described, but the present invention is not limited thereto. FIG. 6 is a schematic diagram showing a configuration of a transcoding system 600 according to a modification. The transcoding system 600 includes a decoding device 602 and an encoding device 604 that is separate from the decoding device 602. The decoding device 602 includes a decoding unit 102, a resolution conversion unit 146, and an acquisition unit 140. The encoding device 604 includes a determination unit 142, an encoding control unit 144, and an encoding unit 104. The decoding device 602 and the encoding device 604 are connected by wire (for example, HD-SDI) or wirelessly. In this example, the decoding device 602 and the encoding device 604 are connected by an uncompressed video transmission interface 606. The decoding device 602 acquires the input bitstream BSI. The reconstructed moving image after resolution conversion is transmitted from the decoding device 602 to the encoding device 604 as the video signal SMP via the uncompressed video transmission interface 606. The encoding device 604 outputs an output bit stream BST.

  The decoding device 602 superimposes the motion information, the motion vector, the compression mode, the presence / absence of an error signal, and the decoded quantization value acquired by the acquisition unit 140 on the video signal SMP transmitted through the uncompressed video transmission interface 606 ( Alternatively, it is embedded in the video signal SMP). The encoding device 604 extracts various superimposed information from the video signal SMP received through the uncompressed video transmission interface 606 (or extracts it from the video signal SMP). As an embedding method, for example, LSB of a luminance component may be used. Although uncompressed video is inherently lossless, since the reconstructed image obtained by decoding the input bitstream BSI originally contains coding noise, the influence of embedding data necessary for coding is small. According to this modification, it is not necessary to provide a separate cable or the like for transmitting the information acquired by the acquisition unit 140, so that the transcoding system 600 can be configured more simply.

  10 cable TV system, 12 broadcaster server, 14 center equipment, 16 cable network, 18 end user, 100 transcoding device.

Claims (11)

First acquisition means for acquiring a bit stream of a moving image;
Output means for outputting a new bitstream having a different setting from the acquired bitstream;
Decoding means for entropy decoding the acquired bitstream;
Second acquisition means for acquiring, for each block, a motion vector, an amount of an error signal, and a motion vector compression mode from the result of entropy decoding;
First determining whether to apply the acquired motion vector to a corresponding encoding target block when generating the new bitstream based on the acquired amount of the error signal A determination means;
Second determination means for determining whether to apply the acquired compression mode to the block to be encoded when generating the new bitstream based on the acquired compression mode; Transcoding system comprising. 2. The transcoding system according to claim 1, further comprising a prediction unit that performs inter prediction on the block to be encoded using the acquired motion vector when it is determined to be applied by the first determination unit. . 3. The transcoding system according to claim 1, further comprising a control unit that deactivates a motion search for the block to be encoded when it is determined to be applied by the first determination unit .   4. The transcoding system according to claim 1, wherein the first determination unit determines to apply when the amount of the acquired error signal is zero and determines not to apply otherwise. 5. . 5. The transcoding system according to claim 1, wherein the second determination unit determines to apply when the acquired compression mode is a mode in which a motion vector of another block is reused . 6. The second acquisition means further acquires a quantization value for each block from the result of entropy decoding,
6. The transcoding system according to claim 1, further comprising a setting unit configured to set a lower limit of a quantization value in quantization of the block to be encoded based on the acquired quantization value. The transcoding system described in 1. The transcoding system according to claim 6, wherein the setting means sets the lower limit based on a ratio between the resolution of the acquired bit stream and the resolution of the new bit stream and the acquired quantization value. . A decoding device and an encoding device separate from the decoding device;
The decoding device includes the first acquisition unit, the decoding unit, and the second acquisition unit,
The encoding device includes the output means, the first determination means, and the second determination means,
The decoding device and the encoding device are connected by an uncompressed video transmission interface, and the decoding device superimposes information acquired by the second acquisition means on a video signal of the uncompressed video transmission interface, and the encoding The transcoding system according to claim 1, wherein the apparatus extracts information superimposed on the video signal . Getting a video bitstream,
Outputting a new bit stream having a different setting from the acquired bit stream;
Entropy decoding the acquired bitstream;
Obtaining the motion vector, the amount of error signal, and the motion vector compression mode for each block from the result of entropy decoding;
Determining whether to apply the acquired motion vector to a corresponding block to be encoded when generating the new bitstream based on the amount of the acquired error signal; ,
Determining whether or not to apply the acquired compression mode to the block to be encoded when generating the new bitstream based on the acquired compression mode. Code method . A computer-readable recording medium comprising a program that causes a computer to function as the transcoding system according to claim 1. An encoding apparatus comprising the output unit, the first determination unit, and the second determination unit included in the transcoding system according to any one of claims 1 to 8.

Images