JP2009118233A - Moving image encoding apparatus and decoding apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a moving image encoding apparatus and decoding apparatus which adaptively suppress encoding side information to thereby reduce the amount of codes which are not effectively utilized in an encoding syntax. <P>SOLUTION: The moving image encoding apparatus receives encoding mode restriction information and intra-prediction direction restriction information. In accordance with the restriction information, an extension syntax description section generates bit stream data 232a as illustrated. Namely, the data 232a are constituted of a sequence parameter set (SPS), a picture parameter set (PPS), a slice header and slice data, the picture parameter set (PPS) is constituted of a conventional H.264 syntax description and an additional component of the present invention, and the slice header is constituted of the conventional H.264 syntax description and an additional component of the present invention, wherein the additional component of the present invention is described either in the picture parameter set (PPS) or in the slice header. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a video encoding device and a decoding device, and in particular, when encoding in macroblock (MB) units, by applying an optimal one from a plurality of candidates for a coding mode and an intra prediction direction, The present invention relates to a moving picture coding apparatus and a decoding apparatus that control coding modes and intra prediction directions to be applied and suppress coding side information adaptively in a moving picture coding apparatus capable of obtaining high coding efficiency. .

  In Non-Patent Document 1, a plurality of encoding modes are defined for encoding in MB units. In the moving image encoding apparatus based on Non-Patent Document 1, an appropriate encoding mode is selected according to video characteristics. High encoding efficiency can be obtained by encoding. In Non-Patent Document 1, a plurality of intra-prediction directions are defined for intra-prediction value generation in intra-coding. In the video encoding device based on Non-Patent Document 1, appropriate intra prediction is performed according to video characteristics. By selecting a direction and encoding, high encoding efficiency can be obtained.

As for the encoding mode and intra prediction direction defined in Non-Patent Document 1, all kinds of encoding modes and intra prediction directions are applied so as to be adaptable to various video features.
Joint Video Team (JVT) of ISO / IEC MPEG and ITU-VCEG, “Text of ISO / IEC 14496 10 Advanced Video Coding 3rd Edition”, July 2004.

  In Non-Patent Document 1, a wide variety of encoding modes and intra prediction directions are defined so as to correspond to various video features. However, in encoding of a certain video, the encoding mode and intra prediction direction that are actually applied are generally limited, and not all encoding modes and intra prediction directions are used. On the other hand, the moving picture encoding apparatus based on Non-Patent Document 1 is designed to generate encoded data even if all encoding modes and intra prediction directions occur in the encoding syntax. Therefore, in most cases, a code amount that is not effectively used is generated for the encoded data and the intra prediction direction in the encoding syntax. The code amount is redundant in encoding and there is room for reduction.

  The present invention has been made in view of the above-described problems of the prior art, and an object of the present invention is to reduce the amount of code that is not effectively used in the coding syntax by adaptively suppressing the coding side information. Another object is to provide a moving picture encoding apparatus and decoding apparatus.

  In order to achieve the above-described object, the present invention provides an encoding mode restriction determination unit in a moving picture encoding apparatus that adaptively uses an encoding mode and an intra prediction direction when encoding in units of macroblocks. An intra prediction direction restriction determination unit, a mode determination control unit that restricts the coding mode used for coding according to the coding mode restriction information, and an intra prediction direction used for coding according to the intra prediction direction restriction information A first feature is that it includes an intra-coded prediction value generation unit and a syntax description unit that changes the syntax description method according to the restriction information on the encoding mode and the intra prediction direction.

  Further, the present invention provides a syntax description unit that changes a syntax description method according to the encoding mode and intra prediction direction restriction information, and the syntax of encoded data when the encoding mode restriction is applied. The second feature is that information indicating that the index table of the coding mode is changed and index information of the coding mode after the change are described.

  Further, the present invention provides a syntax description unit that changes a syntax description method according to the encoding mode and intra prediction direction restriction information, and the syntax of encoded data when the intra prediction direction restriction is applied. 3 describes the information indicating that the index table in the intra prediction direction is changed and the index information in the intra prediction direction after the change.

  Further, the present invention provides a moving picture decoding apparatus for decoding encoded data encoded by the moving picture encoding apparatus, wherein at least a coding mode and an intra prediction direction are analyzed with respect to analysis of syntax description from the received encoded data. It is determined whether one of them is restricted, a restriction information analysis unit that outputs control information, and selected according to the control information output from the restriction information analysis unit, analyzes the encoded data according to the syntax, and encodes First and second syntax analysis units that extract residual signals and prediction information from data, and decoded video based on the residual signals and prediction information output from the first and second syntax analysis units. And a first syntax analysis unit configured to generate an encoding data according to a syntax in which the encoding mode and the intra prediction direction are not limited. Analyzes data, the second syntax analyzer is a fourth feature in that analyzing the coded data in accordance with the syntax of at least one of the encoding mode and the intra prediction directions is limited.

  In the syntax for describing the encoded data, a code amount corresponding to the number of candidates is required with respect to the encoding mode and the intra prediction direction. In the video encoding apparatus of the present invention, only the encoding mode and intra prediction direction suitable for video features are used as candidates, and the encoding mode and intra prediction direction that do not affect encoding (do not contribute to improvement in prediction performance) are Not a candidate. As a result, it is possible to reduce the amount of information necessary for encoding each information regarding the encoding mode and the intra prediction direction. This also makes it possible to reduce the amount of codes with almost no decrease in SNR. That is, the encoding efficiency can be improved.

  In the video decoding device of the present invention, a decoded video can be generated for the encoded data encoded by the video encoding device.

  Hereinafter, the present invention will be described in detail with reference to the drawings. Since the present invention is premised on the video encoding device (Japanese Patent Application No. 2007-85149) previously filed by the present inventors, the video encoding device is first shown in the block diagram of FIG. Will be described with reference to FIG.

  The moving image coding apparatus includes an intra-coded prediction value generation unit 1, an inter-coding prediction value generation unit 2, a mode determination control unit 3, a DCT / quantization unit 4, an inverse quantization / inverse DCT unit 5, and a syntax description Section 6, entropy encoding section 7, local memory (1) 8, and local memory (2) 9.

  The intra-coded prediction value generation unit 1 receives the pixel value of the input video a, the prediction information c in the encoded MB, and the locally decoded pixel value b in the encoded MB. Also, as the output of the intra-coded prediction value generation unit 1, the finally selected intra prediction direction e, the intra prediction value d in the intra prediction direction, and the cost value f in the intra prediction direction are output.

  A detailed configuration of the intra-coded prediction value generation unit 1 will be described with reference to the block diagram of FIG. As illustrated, the intra-coded prediction value generation unit 1 includes a prediction value generation unit 11, a distortion calculation unit 12, a code amount calculation unit 13, and a prediction direction determination unit 14.

  The prediction value generation unit 11 receives the pixel value b that is locally decoded in the encoded block as an input, and calculates the prediction value 11a from the input pixel value a based on the intra prediction direction c defined in H.264. Generate. As an output of the prediction value generation unit 11, a prediction value 11a based on each prediction direction is output. In addition, the residual signal 12a regarding each prediction direction is obtained by calculating | requiring the difference of the predicted value 11a which is the output of this predicted value production | generation part 11 with respect to the pixel value a of an input image | video.

  The distortion calculation unit 12 receives the residual signal 12a for each prediction direction, and obtains a value (encoding distortion amount) 12b obtained by summing the square of the residual signal in each pixel in each block for each prediction direction. Then, as the output of the distortion calculation unit 12, the encoded distortion amount 12b for each prediction direction is output.

  In the code amount calculation unit 13, the residual signal 12a for each prediction direction and the intra prediction direction c selected in the encoded block adjacent to the processing block are input, and the input residual signal 12a for each prediction direction. From this, the code amount 13b required for encoding the residual signal is obtained. For each prediction direction, the code amount 13b required for encoding in the prediction direction is obtained from the intra prediction direction c in the input encoded block. After the calculation process, for each prediction direction, a total of the residual signal and the amount of code required for encoding the prediction direction is obtained. As the output of the code amount calculation unit 13, the code amount 13b required for encoding is output for each prediction direction.

  For each prediction direction, the prediction direction determination unit 14 receives the encoding distortion amount 12b, the intra prediction value, and the code amount 13b required for encoding. The prediction direction determination unit 14 calculates a cost value from the coding distortion amount and the code amount for each intra prediction direction. Furthermore, the magnitude relations of the cost values in each prediction direction are compared, the prediction direction having the smallest cost value is obtained, and the prediction direction is regarded as the optimum prediction direction. As the output of the prediction direction determination unit 14, the prediction value d in the optimal prediction direction, the optimal prediction direction e, and the cost value f in the optimal prediction direction are output.

  Referring back to FIG. 1, the inter-coded prediction value generation unit 2 continues with the pixel value a of the input video, the locally decoded pixel value b in the encoded frame, and the encoded block adjacent to the processing block. And an appropriate motion vector is obtained from the pixel value a of the input video and the locally decoded pixel value b in the encoded frame. Further, a predicted value and a residual signal are obtained based on the motion vector. Further, the coding distortion amount and the code amount required for inter coding are calculated from the residual signal, and the cost value is obtained from the residual signal and the code amount. From the above result, the inter-coded prediction value generation unit 2 outputs, as its output, the prediction value g for the appropriate motion vector, the appropriate motion vector h, and the cost value i for the motion vector h.

  In the mode determination control unit 3, the cost value f in the intra prediction output from the intra-coded prediction value generation unit 1 and the cost value i in the inter-prediction output from the inter-coding prediction value generation unit 2 are input. The cost value in prediction and inter prediction is compared, and prediction with a small cost value is set as prediction in the block. As an output of the mode determination control unit 3, control information j for switching between intra coding and inter coding is output.

  The DCT / quantization unit 4 receives a residual signal k, which is a difference between a prediction value and a pixel value a of the input video. The DCT / quantization unit 4 performs DCT processing on the residual signal k. Also, quantization is performed on the DCT coefficient obtained by the DCT processing. As an output of the DCT / quantization unit 4, a quantized DCT coefficient l is output.

  The inverse quantization / inverse DCT unit 5 receives the quantized DCT coefficient 1 as an input. The inverse quantization / inverse DCT unit 5 performs inverse quantization on the quantized DCT coefficient l. Also, inverse DCT is performed on the result of inverse quantization. As an output of the inverse quantization / inverse DCT unit 5, a result obtained by applying the inverse DCT is output. Note that a locally decoded pixel value b is generated by adding a prediction value to the inverse DCT result output from the inverse quantization / inverse DCT unit 5.

  In the syntax description section 6, the quantized DCT coefficient l and the prediction method (intra prediction direction for intra coding, motion vector for inter coding) c used in the block are input. The syntax description unit 6 determines the contents to be described in the syntax according to the syntax description method defined in H.264 for the information on the quantized DCT coefficient and the prediction direction. Specifically, parameters for each syntax are determined. As the output of the syntax description unit 6, the setting value information m for each syntax parameter is output.

  In the entropy encoding unit 7, setting value information m for each syntax parameter output from the syntax description unit 6 is input, entropy encoding is performed according to the setting value in each syntax parameter, and encoded data is obtained. Generate. Then, the encoded stream n is output as the output of the entropy encoding unit 7.

  In the local memory (1) 8, the locally decoded pixel value b is input, the locally decoded pixel value b is accumulated, and the pixel value is appropriately converted into an intra-coded prediction value generation unit as necessary. 1 and the inter-coded prediction value generation unit 2.

  In the local memory (2) 9, information c and c ′ relating to intra prediction (intra prediction direction) and inter prediction (motion vector) are input, information c and c ′ relating to the prediction are accumulated, and if necessary, As appropriate, the intra prediction direction c is supplied to the intra-coded prediction value generation unit 1 and the motion vector c ′ is supplied to the inter-coding prediction value generation unit 2.

  FIG. 3 shows a block diagram of a video decoding device corresponding to the video encoding device. The moving picture decoding apparatus includes an entropy decoding unit 31, a syntax analysis unit 32, a prediction method determination unit 33, an intra prediction value generation unit 34, an inter prediction value generation unit 35, an inverse DCT / inverse quantization unit 36, and a memory 37. Is done.

  The entropy decoding unit 31 receives an entropy-encoded encoded stream 31a and performs entropy decoding on the input encoded stream. Then, the entropy-decoded encoded data 31b is output as an output of the entropy decoding unit 31.

  The syntax analysis unit 32 receives encoded data as an input, The syntax analysis is performed according to the syntax defined in H.264. As a result of the analysis, a residual signal 32a and prediction information 32b are extracted from the encoded data, and the residual signal 32a and prediction information 32b are output.

  The prediction method determination unit 33 inputs prediction information and determines a prediction method for the MB for the prediction information. Then, as the output of the prediction method determination unit 33, information for controlling intra coding and inter coding is output.

  The intra prediction value generation unit 34 receives the decoded pixel value and the prediction information (intra prediction direction) as inputs, and generates an intra prediction value using the decoded pixel value according to the prediction information. As the output, an intra prediction value is output. The inter prediction value generation unit 35 receives the decoded pixel value and the prediction information (motion vector) as inputs, and generates an inter prediction value using the decoded pixel value according to the prediction information. As the output, an inter prediction value is output.

  In the inverse DCT / inverse quantization unit 36, the quantized DCT coefficient related to the residual signal obtained from the encoded data is input, and inverse quantization and inverse DCT are performed on the quantized DCT coefficient. . The inverse DCT / inverse quantization unit 36 outputs the inverse DCT value as its output. Note that the decoded video 37a is obtained by adding the output of the inverse DCT / inverse quantization unit 36 and the prediction value (intra prediction value or inter prediction value).

  In the memory 37, the pixel value of the decoded video 37 a is input, the decoded pixel value is accumulated, and the decoded pixel value is appropriately supplied to the intra prediction value generation unit 34 and the inter prediction value generation unit 35.

  The above is the description of the moving picture encoding apparatus and the moving picture decoding apparatus which are the premise of the present invention. The present invention will be described in detail below.

  The moving picture coding apparatus according to the present invention, in moving picture coding that adaptively uses a coding mode and an intra prediction direction at the time of MB unit coding, a coding mode and an intra that are candidates according to video characteristics. By controlling the prediction direction, the amount of code for encoding the information is reduced. Specifically, in encoding in MB units, encoding mode candidates are selected in advance, and only encoding modes suitable for video features are used as candidates. Further, in intra coding, the intra prediction direction is selected in advance, and only the intra prediction direction suitable for the video feature is set as a candidate.

  FIG. 4 is a block diagram showing a configuration of a moving image encoding apparatus according to an embodiment of the present invention. In the figure, the moving picture coding apparatus includes an intra-coded prediction value generation unit 21, an inter-coding prediction value generation unit 2, a mode determination control unit 22, a DCT / quantization unit 4, an inverse quantization / inverse DCT unit 5, A syntax description unit 23, an entropy encoding unit 7, a local memory (1) 24, a local memory (2) 9, an encoding mode restriction determination unit 25, and an intra prediction direction restriction determination unit 26 are included. In FIG. 4, the same reference numerals as those in FIG. 1 perform the same functions.

  In the present embodiment, as will be apparent from the following description, it is possible to arbitrarily provide restriction information regarding the encoding mode and the intra prediction direction. That is, it is possible to adaptively suppress the encoded side information. The restriction information regarding the encoding mode and the intra prediction direction describes information regarding the encoding mode and the intra prediction direction that can be used for encoding.

  The encoding mode restriction determination unit 25 receives as input the pixel value a of the input video and the locally decoded pixel value b in the encoded frame. Further, when the restriction information 25a regarding the coding mode is given from the outside as an input to the moving picture coding apparatus, the information 25a is used as an input. When the restriction information 25a is given, the coding mode restriction determination unit 25 determines a coding mode applicable at the time of coding according to the restriction information 25a. When the restriction information 25a is not given, an encoding mode applicable at the time of encoding is determined from the pixel value a of the input video and the locally decoded pixel value b. As an output of the encoding mode restriction determination unit 25, information (encoding mode restriction information) 25b relating to an encoding mode applicable at the time of encoding is output.

  The intra prediction direction restriction determination unit 26 receives as input the pixel value a of the input video and the locally decoded pixel value b in the encoded frame. In addition, when the restriction information 25a related to the intra prediction direction is given from the outside as an input to the moving picture coding apparatus, the restriction information 25a is input, and the intra applicable for encoding according to the restriction information 25a. The prediction direction 26b is determined. If the restriction information 25a is not given, an intra prediction direction 26b applicable at the time of encoding is determined from the pixel value a of the input video and the locally decoded pixel value b. As an output of the intra prediction direction restriction determination unit 26, information (intra prediction direction restriction information) 26b regarding the intra prediction direction applicable at the time of encoding is output.

  FIG. 5 is a block diagram showing a detailed configuration of the intra-coded prediction value generation unit 21. The intra-coded prediction value generation unit 21 includes a prediction value generation unit 11, a distortion calculation unit 12, a code amount calculation unit 13, and a prediction direction determination unit 211. However, since the predicted value generation unit 11, the distortion calculation unit 12, and the code amount calculation unit 13 have the same functions as the same unit in FIG.

  The prediction direction determination unit 211 receives the intra prediction direction restriction information 26b and the encoding distortion amount 12b, the intra prediction value 11a, and the code amount 13b required for encoding for each prediction direction. The prediction direction determination unit 211 calculates a cost value f from the coding distortion amount 12b and the code amount 13b for each intra prediction direction. Further, the magnitude relations of the cost values in the respective prediction directions are compared, the prediction direction e having the smallest cost value is obtained, and the prediction direction is regarded as the optimum prediction direction e. However, the prediction direction for comparing the magnitude relation of the cost value f is limited to the prediction direction that can be applied in the intra prediction direction restriction information 26b. As the output of the prediction direction determination unit 211, the prediction value d in the optimal prediction direction, the optimal prediction direction e, and the cost value f in the optimal prediction direction are output.

  Returning to FIG. 4, the description will be continued. In the mode determination control unit 22, the cost value f in the intra prediction output from the intra coded prediction value generation unit 21, the cost value i in the inter prediction output from the inter coding prediction value generation unit 2, and the coding mode restriction Information 25b is input. The mode determination control unit 22 compares the cost values f and i in the intra prediction and the inter prediction, and sets the prediction having a small cost value as the prediction in the block. However, when an applicable encoding mode is limited in the encoding mode restriction information 25b, the applicable encoding mode is preferentially used for prediction in the block. As the output of the mode determination control unit 22, control information j for switching between intra coding and inter coding is output.

  In the local memory (1) 24, the locally decoded pixel value b is input, the locally decoded pixel value is accumulated, and if necessary, the pixel value b is appropriately converted into the intra-coded prediction value generation unit 21, The inter-coded prediction value generation unit 2, the coding mode restriction determination unit 25, and the intra prediction direction restriction determination unit 26 are supplied. Note that the intra-coded prediction value generation unit 21 is supplied with the locally decoded pixel value b in the encoded block adjacent to the processing block, and the inter-coding prediction value generation unit 2 and the coding mode restriction determination unit 25. The intra-predicted direction restriction determination unit 26 is supplied with the locally decoded pixel value b in the encoded frame.

  FIG. 6 shows a block diagram of a detailed configuration of the syntax description unit 23. The syntax description unit 23 includes an H.264 syntax description unit 231, an extended syntax description unit 232, and a syntax description control unit 233. However, the H.264 syntax description part 231 has the same function as the syntax description part in the conventional method.

  In the extended syntax description unit 232, prediction information c in encoding (intra prediction direction in the case of intra encoding, motion vector in the case of inter encoding), residual signal l, encoding mode restriction information 25b, intra prediction The direction restriction information 26b is input. The extended syntax description unit 232 determines the contents to be described in the syntax according to the syntax description method defined in H.264 for the quantized DCT coefficient l. In addition, for information regarding the coding mode in the block, the index representing the coding mode is reconstructed according to the coding mode restriction information 25b, and the contents to be described in the syntax are determined according to the reconstructed index. Similarly, with respect to the information regarding the intra prediction direction in the block, the index representing the intra prediction direction is reconstructed according to the intra prediction direction restriction information 26b, and the contents to be described in the syntax are determined according to the reconstructed index. As the output of the extended syntax description section 232, the setting value information 232a for each syntax parameter is output.

  The syntax description control unit 233 receives the encoding mode restriction information 25b and the intra prediction direction restriction information 26b as input, and relates to the coding mode and the intra prediction direction from the coding mode restriction information 25b and the intra prediction direction restriction information 26b. Determine whether the proposed method is applied. When the proposed method is applied in either case, control is performed so that the encoded data 232a is obtained from the extended syntax description unit 232 with respect to the syntax description, and when only the conventional method is applied in either case, H Control is performed so that encoded data 231a is obtained from the .264 syntax description unit 231. In the encoded data 231a and the encoded data 232a, an identification code is added or inserted, and an encoding mode set (encoding mode and intra prediction direction) to be applied is designated.

  Next, an outline of the encoded data (encoded stream data) 232a output from the extended syntax description unit 232 will be described with reference to FIG. The encoded stream data 232a includes a sequence parameter set (SPS), a picture parameter set (PPS), a slice header, and slice data. According to the present invention, as shown in the figure, a picture parameter set (PPS) is composed of a conventional H.264 syntax description and an addition according to the present invention, and a slice header is composed of a conventional H.264 syntax description and a main part. The additional portion according to the present invention is described in one of the picture parameter set (PPS) and the slice header.

  Here, the additions according to the present invention include, for example, parameters (1) coding_mode_table_update, (2) table_update_I_pic, (3) table_update_P_pic, (4) table_update_B_pic, (5) coding_mode_I_table [i], (6) coding_mode_P_table [i], ( 7) It consists of coding_mode_B_table [i], (8) intra4x4_predict_table_update, (9) intra8x8_predict_table_update, (10) intra4x4_predict_table [i], and (11) intra8x8_predict_table [i]. The parameters (1) to (7) are parameters related to the encoding mode restriction, and the parameters (8) to (11) are parameters related to the intra prediction direction restriction.

  Here, the parameters (1) to (11) will be described as follows.

  (1) coding_mode_table_update A flag indicating that the index table is changed with respect to information representing the coding mode. If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set.

  (2) When table_update_I_pic coding_mode_table_update is 1, a flag indicating that the index table is changed with respect to information representing the coding mode in the I picture (Intra16x16). If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set.

  (3) When table_update_P_pic coding_mode_table_update is 1, a flag indicating that the index table is changed with respect to information representing the coding mode in the P picture. If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set.

  (4) When table_update_B_pic coding_mode_table_update is 1, a flag indicating that the index table is changed with respect to information representing the coding mode in the B picture. If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set.

  (5) coding_mode_I_table [i] When table_update_I_pic is 1, this parameter indicates the change of the index table for information expressing the coding mode in I picture (Intra16x16). i is the index number after the change, and the encoding mode index number in the conventional method corresponding to the encoding mode assigned to the index number is set as a value given to the parameter.

  (6) coding_mode_P_table [i] When table_update_P_pic is 1, this parameter indicates the change of the index table regarding information expressing the coding mode in the P picture. i is the index number after the change, and the encoding mode index number in the conventional method corresponding to the encoding mode assigned to the index number is set as a value given to the parameter.

  (7) coding_mode_B_table [i] When table_update_B_pic is 1, this parameter indicates the change of the index table for information expressing the coding mode in B picture. i is the index number after the change, and the encoding mode index number in the conventional method corresponding to the encoding mode assigned to the index number is set as a value given to the parameter.

  Note that the same effect can be obtained by replacing 0 and 1 with respect to (1) coding_mode_table_update, (2) table_update_I_pic, (3) table_update_P_pic, and (4) table_update_B_pic.

  (8) intra4x4_predict_table_update A flag indicating that the index table is changed with respect to the intra prediction direction of Intra4x4. If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set. The same effect can be obtained even if 0 and 1 are interchanged.

  (9) intra8x8_predict_table_update A flag indicating that the index table is changed with respect to the intra prediction direction of Intra8x8. If 0, the index table is not changed, and 1 indicates that the index table is changed. If the same parameter does not exist, 0 is set. The same effect can be obtained even if 0 and 1 are interchanged.

  (10) intra4x4_predict_table [i] A parameter that represents a change in the index table regarding the intra prediction direction of Intra4x4. i is the index number after the change, and an intra prediction direction index number in the conventional method corresponding to the intra prediction direction assigned to the index number is set as a value given to the parameter.

  (11) intra8x8_predict_table [i] A parameter indicating a change in the index table with respect to the intra prediction direction of Intra8x8. i is the index number after the change, and an intra prediction direction index number in the conventional method corresponding to the intra prediction direction assigned to the index number is set as a value given to the parameter.

  Next, using the above-described method for generating encoding mode restriction information according to video characteristics, a method for generating intra prediction direction restriction information according to video characteristics, and the encoding mode restriction information and intra prediction direction restriction information A specific example of syntax description in the case of encoding will be described below.

  <1> Method for creating encoding mode restriction information

  A description will be given of a method for selecting only a coding mode according to video characteristics as a candidate for selection of a coding mode in MB unit coding. There are two methods for restricting encoding mode candidates: (a) a method of inputting the encoding mode restriction information 25a or 25b from the outside; and (b) encoding mode restriction information based on video features and encoded information. This is a method for determining 25a or 25b.

  As for (a), according to the input encoding mode restriction information, the encoding mode that is a candidate for encoding in MB units is controlled. As a method for generating the encoding mode restriction information, a method is conceivable in which an encoding process based on a conventional method is performed on a processing frame, and only a higher-order encoding mode with a high occurrence rate is a candidate.

  For (b), first, regarding the processing frame and the immediately preceding frame of the same picture type, whether the video characteristics of both frames are similar from the average value of pixel values, the dispersion value of pixel values, and the global motion between frames Determine whether. When it is determined that the video features of both frames are similar, the encoding mode in the immediately preceding same picture type is followed.

  Also, encoding mode restriction information is generated by determining the encoding mode division size from the pixel value variance in the processing frame, and determining whether or not intra-coding can be applied from the global motion of the processing frame and the immediately preceding frame. Is possible.

  Furthermore, using the processing frame, the encoded local decoded pixel value, and the encoded prediction information, a method for appropriately and quickly estimating the encoding mode in the processing frame is, for example, the following [2] to [ 4] etc., as shown in the literature. By using the methods shown in these documents, the encoding mode restriction information for estimating the encoding mode in the processing frame at high speed and using only the top of the encoding mode that is estimated to be high in frequency as the candidate is displayed. It is also possible to generate.

[2] Xuan Jing and Lap-Pui Chau, “An efficient inter mode decision approach for H.264 video coding,” IEEE ICME2004, pp.1111-1114, 2004.
[3] Hua Wang and Xiangjun Zhu , "Adaptive motion estimation and mode decision strategy for H.264 encoder," Proceedings of the 6 th World Congress on Intelligent Control and Automation, pp.10323-10327, June 2006.
[4] Andy C. Yu and Graham R. Martin, "Advanced block size selection algorithm for inter frame coding in H.264 / MPEG-4 AVC," ICIP2004, pp.95-98, 2004.

  When encoding is performed based on the encoding mode restriction information generated by the above (a) and (b), a change from the conventional method is required regarding the syntax description of the encoded data. The content of the syntax change is the addition of index table information related to the encoding mode.

  Regarding the addition of the index table related to the encoding mode, as already described with reference to FIG. 7, an index table is assigned in units of picture parameter sets (PPS) or slices. That is, the “addition according to the present invention” is added to the conventional technology (conventional description) with respect to the syntax definition of either the PPS or the slice header. For the syntax definition of PPS and slice header in the conventional description, see Document [1].

  A specific example of syntax description in the method of the present invention will be described below. The following specific example shows an example in which it is determined that the 8 × 16 coding mode is effective as a coding mode candidate in the B picture.

  The 8 × 16 encoding mode expression in the B picture by the conventional method is as shown in FIG. 8, and these are expressed by 5 bits. Therefore, the parameter (1) coding_mode_table_update is set to 1, (4) table_update_B_pic is set to 1, and (7) coding_mode_B_table [i] is set to [0] = 0 by the encoding mode restriction information according to the video feature according to the present invention. , [1] = 5, [2] = 7, [3] = 9, [4] = 13, [5] = 15, [6] = 17, [7] = 19, [inferred] = inferred If this is the case, the encoding mode is limited as shown in FIG. 9 and is expressed by 3 bits. In this case, the syntax relating to the encoding mode can be reduced by 2 bits.

  <2> Method for creating intra prediction direction restriction information

  Regarding the selection of the intra prediction direction in intra coding, a method of selecting only the intra prediction direction corresponding to the video feature is shown. There are two methods for restricting intra prediction direction candidates: (a) a method of inputting intra prediction direction restriction information 26a or 26b from the outside, and (b) intra prediction direction restriction information from video features and encoded information. This is a method for determining 26a or 26b.

For (a), the intra prediction direction in intra coding is limited according to the input intra prediction direction restriction information. As a method for generating intra prediction direction restriction information, a method is conceivable in which intra coding processing based on a conventional method is performed on a processing frame, and only a higher rank in the intra prediction direction having a high occurrence frequency is used as a candidate.
For (b), first, determine whether the video features of both frames are similar from the average value of pixel values, the variance value of pixel values, and the global motion between frames for the processing frame and the frame of the previous intra picture. To do. When it is determined that the video features of both frames are similar, the intra prediction direction in the immediately preceding intra picture is followed. Further, it is possible to generate intra prediction direction restriction information by determining the intra prediction direction from the global motion with the processing frame and the immediately preceding frame.
Further, a technique for appropriately and quickly estimating the intra prediction direction in the processing frame using the processing frame, the encoded locally decoded pixel value, and the encoded prediction information is described in the following documents [5] [6] ], Many have been proposed. By using the methods shown in these documents, intra prediction direction restriction information for which the intra prediction direction in the processing frame is estimated at high speed and only the upper rank of the intra prediction direction that is estimated to be high in frequency is used as a candidate. It is also possible to generate.

[5] Chao-Hsuing Tseng, Hung-Ming Wang, and Jar-Ferr Yang, “Enhanced intra-4x4 mode decision for H.264 / AVC coders,” IEEE CSVT 2006, Vol. 16, No. 8, August 2006.
[6] Zhi-Yi Mai, Chun-Ling Yang and Sheng-Li Xie, “Improved best prediction mode selection methods based on structural similarity in H.264 I-frame encoder,” Systems, Man and Cybernetics, 2005.

  When encoding is performed based on the intra prediction direction restriction information generated by (a) and (b) above, a change from the conventional method is required regarding the syntax description of the encoded data. The content of the syntax change is the addition of index table information related to the intra prediction direction and the reduction of the number of bits used for encoding the intra prediction direction.

  Regarding the addition of the index table related to the intra prediction direction, as already described with reference to FIG. 7, the index table is assigned in units of picture parameter sets (PPS) or slices. That is, the “addition according to the present invention” is added to the conventional technology (conventional description) for the syntax definition of the PPS and the slice header. However, see [1] for the syntax definition of PPS and slice header in the conventional description.

  For the reduction of the number of bits used for encoding the intra prediction direction, the minimum number of bits that can represent the intra prediction direction after changing the index table is used. That is, the intra prediction direction is represented by 1 bit when the number of intra prediction directions is 2 or less, 2 bits when 3 or more and 5 or less, and 3 bits when 6 or more.

  A specific example of syntax description in the method of the present invention will be described below. In this specific example, as an intra prediction direction candidate in Intra4 × 4 or Intra8 × 8, only three directions of horizontal prediction, DC prediction, and horizontal upward prediction are shown.

  That is, the intra direction prediction in the conventional method is as shown in FIG. 10, and these are expressed by 3 bits. Therefore, the above parameter (8) intra4x4_predict_table_update is set to 1 by the intra prediction direction restriction according to the video feature according to the present invention, and (10) In intra4x4_predict_table [i], [0] = 1, [1] = 8, [ 2] = 2, the intra direction prediction is limited as shown in FIG. 11 and is expressed by 2 bits. In this case, the syntax for intra direction prediction can be reduced by 1 bit.

  Next, an embodiment of the video decoding device of the present invention will be described with reference to the block diagram of FIG. The moving picture decoding apparatus according to the present embodiment is obtained by replacing the syntax analysis unit 32 of FIG. 3 with the configuration of FIG. 12 and other configurations are the same as those of FIG. Only the description part will be described, and description of other components will be omitted.

  The restriction information analysis unit 323 receives the entropy-decoded encoded data 31b as input, and extracts identification information regarding whether or not to apply the method of the present invention included in the encoded data. Then, it controls which of the conventional method and the method of the present invention is applied to the syntax analysis of the encoded data 31b. As the output of the restriction information analysis unit 323, the control information 323a related to the syntax analysis method for the encoded data 31b is output. According to the control information 323a, the H.264 data Selection (switching) of the input side and output side of the H.264 syntax analysis unit 321 and the extended syntax analysis unit 322 is performed.

  H. In the H.264 syntax analysis unit 321, The encoded data 31b described in accordance with the syntax defined in H.264 is input, the encoded data is analyzed in accordance with the syntax to specify the encoding mode set, and the residual signal 32a and the prediction information 32b are determined from the encoded data. And extract. And this H.I. As an output of the H.264 syntax analysis unit 321, a residual signal 32a and prediction information 32b are output.

  The extended syntax analysis unit 322 receives the encoded data 31b described according to the syntax defined in the method of the present invention, analyzes the encoded data according to the syntax, specifies the encoding mode set, Residual signal 32a and prediction information 32b are extracted from the data. Then, as the output of the extended syntax analysis unit 322, a residual signal 32a and prediction information 32b are output.

  Regarding the encoding mode restriction, the extended syntax analysis unit 322 changes the syntax analysis method of encoded data in accordance with (1) coding_mode_table_update. When the technique of the present invention is applied, the index number of the encoding mode is reconstructed for each picture type according to (2) table_update_I_pic, (3) table_update_P_pic, and (4) table_update_B_pic.

  On the other hand, regarding the intra prediction direction restriction, the extended syntax analysis unit 322 changes the syntax analysis method of encoded data in accordance with (8) intra4x4_predict_table_update and (9) intra8x8_predict_table_update. When the method of the present invention is applied, the index number in the intra prediction direction is reconstructed for each divided block size.

  In order to confirm the effect of the method of the present invention, the present inventor conducted a simulation experiment on the restriction on the intra prediction direction. The proposed method was implemented in JM12.1 for ITE HD standard video (1920 pixels x 1080 lines). In addition, about the restriction | limiting information of an intra prediction direction, the restriction | limiting information shall be input from the outside. In addition, the restriction information is assumed to use the prediction direction in which the frequency of occurrence is the top 3 or the top 5 when encoding using JM12.1 under the same conditions as the experimental conditions. As a result, with respect to the average code amount reduction with respect to the conventional method, a gain of 0.42% was obtained in the case of the top five and 0.52% was obtained in the case of the top three.

It is a block diagram which shows the structure of the moving image encoding apparatus used as the premise of this invention. It is a block diagram which shows the detail of the intra coding prediction value production | generation part of FIG. It is a block diagram which shows the structure of the moving image decoding apparatus used as the premise of this invention. It is a block diagram which shows the structure of the moving image encoder of one Embodiment of this invention. It is a block diagram which shows the detail of the intra coding prediction value production | generation part of FIG. FIG. 5 is a block diagram illustrating details of a syntax description unit in FIG. 4. It is explanatory drawing of the bit stream data which shows the output of the extended syntax description part of FIG. It is explanatory drawing of the conventional encoding mode expression in a B picture. It is explanatory drawing of the example of a syntax description regarding the encoding mode restriction | limiting by this invention. It is explanatory drawing of intra prediction direction expression by a conventional method. It is explanatory drawing of the syntax description example regarding the intra prediction direction restriction | limiting by this invention. It is a block diagram which shows the structure of the syntax analysis part of the moving image decoding apparatus of this invention.

Explanation of symbols

  2 ... Inter-coded prediction value generation unit, 4 ... DCT / quantization unit, 5 ... Inverse quantization / inverse DCT unit, 7 ... Entropy encoding unit, 9 ... Local memory ( 2), 11 ... predicted value generation unit, 12 ... distortion calculation unit, 13 ... code amount calculation unit, 211 ... prediction direction determination unit, 21 ... intra coded prediction value generation unit, 22 ... Mode determination control unit, 23 ... Syntax description unit, 24 ... Local memory (1), 25 ... Coding mode limit determination unit, 26 ... Intra prediction direction limit determination unit, 231 ... H. 264 syntax description part, 232 ... extended syntax description part, 31 ... entropy decoding part, 32 ... syntax analysis part, 33 ... prediction method determination part, 34 ... intra prediction value generation 35, Inter prediction value generation unit, 321 ... H.264 syntax analysis unit, 322 ... Extended syntax analysis unit, 323 ... Restriction information analysis unit.

Claims (10)

In the video encoding apparatus that adaptively uses the encoding mode and the intra prediction direction when encoding in units of macroblocks,
An encoding mode restriction determination unit, an intra prediction direction restriction determination part, a mode determination control unit that restricts an encoding mode used for encoding according to the coding mode restriction information, and an encoding according to the intra prediction direction restriction information An intra-coded prediction value generation unit that limits the intra-prediction direction used in the above, and a syntax description unit that changes the syntax description method according to the encoding mode and the restriction information of the intra-prediction direction,
The syntax description unit reduces a code amount used for encoding information related to the encoding mode and the intra prediction direction. The moving image encoding device according to claim 1,
The coding mode restriction determination unit generates and outputs information for restricting a coding mode used for coding based on restriction information from the outside, or input video and coded data that has been coded. Video encoding device. The moving image encoding device according to claim 2,
The encoding mode restriction determination unit, as a method of restricting an encoding mode used for encoding based on the input video and encoded data, generates an occurrence frequency for an encoding mode applied in an encoded frame. A moving picture encoding apparatus characterized in that, except for the upper part of the processing frame, the processing frame is limited. The moving image encoding device according to claim 1,
The intra prediction direction restriction determination unit generates and outputs information for restricting an intra prediction direction used for intra coding based on restriction information from the outside or input video and encoded data. A moving image encoding device. In the moving image encoding device according to claim 4,
The intra-prediction direction restriction determination unit is a method of restricting an intra-prediction direction used for intra-encoding based on the input video and encoded data. A moving picture encoding apparatus characterized in that, except for a higher rank, a processing frame is limited. The moving image encoding device according to claim 1,
The syntax description unit that changes the syntax description method according to the encoding mode and the intra prediction direction restriction information, when the encoding mode restriction is applied, in the syntax of the encoded data, A moving picture coding apparatus characterized by describing information indicating changing an index table and index information of a coding mode after the change. The moving image encoding device according to claim 1,
The syntax description unit that changes the syntax description method according to the encoding mode and the intra prediction direction restriction information, when the intra prediction direction restriction is applied, A moving picture encoding apparatus characterized by describing information indicating that an index table is changed and index information in the intra prediction direction after the change. In the moving image decoding apparatus which decodes the encoded data encoded by the moving image encoding apparatus according to claim 1,
A restriction information analysis unit that determines whether or not at least one of the coding mode and the intra prediction direction is restricted with respect to the analysis of the syntax description from the received coded data, and outputs control information;
First and second syntax analysis units that are selected according to the control information output from the restriction information analysis unit, analyze the encoded data according to the syntax, and extract the residual signal and the prediction information from the encoded data;
Means for generating a decoded video based on the residual signal output from the first and second syntax analysis units and the prediction information;
The first syntax analysis unit analyzes encoded data according to a syntax in which the encoding mode and the intra prediction direction are not limited, and the second syntax analysis unit analyzes the encoding mode and the intra prediction direction. A moving picture decoding apparatus, wherein encoded data is analyzed according to a syntax in which at least one of them is restricted. The moving picture decoding apparatus according to claim 8, wherein
The second syntax analysis unit specifies an encoding mode set to be applied from the identification code inserted in the encoded data, and extracts an encoding mode or an intra prediction direction in the encoding mode set. A moving picture decoding apparatus characterized by the above. The moving picture decoding apparatus according to claim 8 or 9, wherein
8. A moving picture decoding apparatus that generates decoded video for the encoded data encoded by the moving picture encoding apparatus according to claim 2.

Images