JP2013183207A - Coding device, decoding device, coding method and decoding method of moving image, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a coding device, a decoding device, a coding method and a decoding method for a moving image, and a program, capable of improving coding performance with a suppressed increase of side information transmitted from the moving image coding device to the moving image decoding device.SOLUTION: A moving image coding device AA includes a coding-side memory unit, a converter unit 3, a cost calculator unit 6 and an entropy coder unit 5. The coding-side memory unit stores predetermined basis, and a plurality of suffixes provided to be identifiable of each of the plurality of basis. The converter unit 3 and the cost calculator unit 6 obtain a conversion factor for each of the plurality of basis by applying to each conversion block. Using the applied result, the cost calculator unit 6 determines a base to be applied among the plurality of basis for each conversion block, determines a suffix associated with the determined base as a conversion suffix candidate for each conversion block, and based on the determined result, generates a slice header.

Description

  The present invention relates to a video encoding device, a video decoding device, a video encoding method, a video decoding method, and a program.

  Non-Patent Document 1 discloses one-dimensional DCT (one-dimensional discrete cosine transform) and two-dimensional DCT (two-dimensional discrete cosine transform) as conversion methods for motion compensation residuals. The following two are shown as an outline of the one-dimensional DCT.

  The first is about the autocorrelation of the motion compensation residual. Since the motion compensation residual includes many edge components due to object boundaries and the like, it can be said that the residual signal has a high one-dimensional autocorrelation along the edge. Therefore, when the one-dimensional DCT is applied along the edge, the redundancy can be reduced. However, when the one-dimensional DCT is applied in a direction perpendicular to the edge direction, the signal is dispersed and the redundancy is reduced. Will be increased.

  The second is for applying a one-dimensional DCT along the edge. The edge direction is generally diverse, but here it is assumed that it exists in only one direction within the transform block. For example, assuming that edges exist along the pixels b1 to b4 as shown in FIG. 6A, the pixels are rearranged as shown in FIG. ) Apply one-dimensional DCT in the middle and vertical direction.

  An example of the above-described pixel rearrangement is shown in FIG. Since the rearrangement of these pixels can be realized by using a permutation matrix, a plurality of permutation matrices are defined in advance. Further, a subscript is assigned to each of the plurality of permutation matrices as a unique identifier. When one subscript is determined, one permutation matrix uniquely corresponding to the subscript can be obtained.

  Non-Patent Document 2 shows a standard moving picture coding method, which shows PPS (picture parameter set) as coding information in units of pictures and SH (slice header) as coding information in units of slices. Has been.

F. Kamisli and J. Lim, "Transforms for the motion compensation residual," ICASSP 2009. ISO / IEC 14496-10-MPEG-4 Part 10, Advanced Video Coding.

  According to Non-Patent Document 1, a subscript indicating a permutation matrix used for each transform block is transmitted as side information from the encoding device to the decoding device. For this reason, the code amount can be reduced as compared with the case where the permutation matrix itself is transmitted, but since the subscript indicating the permutation matrix is transmitted for each transform block, the code amount may not be sufficiently reduced.

  Further, in a moving image in which the edge component due to the object boundary is hardly included in the motion compensation residual due to the effect of motion blur, the premise of applying the one-dimensional DCT of Non-Patent Document 1 does not hold, so the two-dimensional DCT is applied. The probability that it will be increased. According to this, the side information increases, and the code amount may increase.

  Therefore, the present invention has been made in view of the above-described problems, and video coding that can improve coding performance by suppressing an increase in side information transmitted from the video coding device to the video decoding device. An object is to provide an apparatus, a moving picture decoding apparatus, a moving picture encoding method, a moving picture decoding method, and a program.

  The present invention proposes the following matters in order to solve the above problems.

  (1) The present invention is a moving image encoding apparatus that encodes a moving image (for example, an input image described later), and can identify a plurality of predetermined bases and each of the plurality of bases. And a plurality of subscripts provided in the encoding side storage means (e.g., corresponding to an encoding side storage unit described later) and each of the plurality of bases is applied to each conversion block to obtain a transform coefficient. Using the application means to be obtained (e.g., equivalent to the conversion unit 3 and the cost calculation unit 6 in FIG. 1) and the application result of the application means, the one to be applied is determined for each conversion block, and determined. Based on the determination result (for example, equivalent to the cost calculation unit 6 in FIG. 1) and a determination result by the determination unit, the encoding information ( example 1 (corresponding to the slice header in FIG. 1) for generating encoded information (for example, equivalent to the cost calculation unit 6 in FIG. 1) and the encoded data (for example, FIG. 1) based on the determination result by the determination unit. Encoding means (e.g., equivalent to the entropy encoding unit 5 in FIG. 1), and the plurality of bases stored in the encoding-side storage means are one-dimensional DCT ( A plurality of types of bases related to a one-dimensional discrete cosine transform) and one type of bases related to a two-dimensional DCT (two-dimensional discrete cosine transform), and the determining means is based on an application result by the applying means. When deciding whether to apply a mixture of two-dimensional DCT and two-dimensional DCT or to apply only two-dimensional DCT, and determining to apply a mixture of one-dimensional DCT and two-dimensional DCT Is determined for each transform block to be applied among the plurality of bases, a subscript associated with the determined base is determined for each transform block as a transform subscript candidate, and only two-dimensional DCT is to be applied In this case, a subscript associated with a base related to the two-dimensional DCT is determined as a conversion subscript candidate for all transform blocks, and the encoding information generation means applies a mixture of the one-dimensional DCT and the two-dimensional DCT. When determined by the determining means, the encoding information including information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied is generated, and the determination means determines that only the two-dimensional DCT is applied. If so, the encoding information including information indicating that only the two-dimensional DCT is applied is generated, and the encoding means includes the one-dimensional DCT and 2D When it is determined by the determining means that the original DCT is applied in a mixed manner, encoded data including a conversion index candidate for each conversion block determined by the determining means and the encoding information is generated, When only the two-dimensional DCT is applied, when it is determined by the determining means, a moving image encoding apparatus is proposed that generates encoded data including the encoding information.

  According to this invention, the moving image coding apparatus for coding a moving image is provided with the encoding side storage unit, the applying unit, the determining unit, the encoded information generating unit, and the encoding unit. Then, the encoding side storage means stores a plurality of predetermined bases and a plurality of subscripts provided so as to be able to identify each of the plurality of bases. The plurality of bases stored in the encoding-side storage means include a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform), a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform), It shall consist of. In addition, the application means determines the transform coefficient by applying it to each transform block for each of the plurality of bases. Further, the decision means decides whether to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner or to apply only the two-dimensional DCT using the application result of the application means. In addition, when the determination unit determines to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner, an applied one of a plurality of bases is determined for each transform block, and a subscript associated with the determined base Is determined for each conversion block as a conversion subscript candidate, and when it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all the conversion blocks. It was decided. In addition, when the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied in a mixed manner by the encoding information generation unit, information indicating that the one-dimensional DCT and the two-dimensional DCT are applied in combination. It was decided to generate the encoded information including. In addition, when only the two-dimensional DCT is applied by the encoding information generation unit, the determination unit determines that the encoding information including information indicating that only the two-dimensional DCT is applied is generated. In addition, when the encoding unit determines that the one-dimensional DCT and the two-dimensional DCT are mixedly applied by the encoding unit, the conversion index candidate for each conversion block determined by the determination unit, the encoding information, When the determination unit determines that only the two-dimensional DCT is applied, the encoded data including the encoding information is generated.

  For this reason, it is determined whether the one-dimensional DCT and the two-dimensional DCT are mixedly applied or only the two-dimensional DCT is applied. When it is determined that the one-dimensional DCT and the two-dimensional DCT are applied together, information indicating that the conversion index candidate determined for each conversion block is applied together with the one-dimensional DCT and the two-dimensional DCT is applied. And encoded data including the encoded information. On the other hand, when it is determined that only the two-dimensional DCT is applied, encoded data including encoded information including information indicating that only the two-dimensional DCT is applied is generated.

  For this reason, when it is determined that only the two-dimensional DCT is applied, the conversion index candidates for each conversion block are not included in the encoded data. Therefore, an increase in side information transmitted from the video encoding device to the video decoding device can be suppressed, and the encoding performance can be improved.

  (2) The present invention relates to the moving picture coding apparatus according to (1), in which the determining means uses a code amount when each of the plurality of bases is applied for each transform block (for example, a generated code amount described later). And a conversion index candidate is determined using the obtained code amount.

  According to the present invention, in the moving picture encoding apparatus according to (1), the determining unit obtains the code amount when each of a plurality of bases is applied for each transform block, and uses the obtained code amount to convert the subscript. The candidate was decided. For this reason, the base to be applied can be determined for each transform block so that the code amount for each transform block is reduced. Therefore, the encoding performance can be further improved.

  (3) The present invention relates to the moving picture encoding apparatus according to (1) or (2), wherein the determining unit performs decoding after applying and encoding each of the plurality of bases for each transform block. Proposing a moving image coding apparatus that generates an image and determines the conversion index candidate using a difference between the decoded image and the moving image (e.g., corresponding to a coding error described later). Yes.

  According to the present invention, in the moving image encoding apparatus according to (1) or (2), the determining unit applies each of a plurality of bases for each transform block, and then decodes to generate a decoded image. Thus, the conversion index candidate is determined using the difference between the decoded image and the moving image. For this reason, the base to be applied can be determined for each transform block so that degradation of the decoded image with respect to the moving image for each transform block is suppressed. Therefore, the encoding performance can be further improved.

  (4) In the moving image encoding device according to any one of (1) to (3), the determination unit determines the conversion subscript candidate among a plurality of types of bases related to the one-dimensional DCT. A subscript that can be determined in advance, based on a result of application by the applying unit, a predetermined one that can be determined as the conversion subscript candidate among a plurality of types of bases related to the one-dimensional DCT, and the two-dimensional DCT Has proposed a moving picture coding apparatus characterized by determining one type of base and a conversion index candidate from among them.

  Here, the same base is applied to the same transform block in the video encoding device that generates the encoded data and the video decoding device that decodes the encoded data. For this reason, the video decoding device needs to be able to recognize all bases that can be applied to each transform block in the video encoding device. Therefore, the moving image encoding apparatus needs to transmit in advance information about all bases that can be applied to each transform block to the moving image decoding apparatus.

  Therefore, according to the present invention, in the moving picture encoding apparatus according to any one of (1) to (3), the determining unit first determines a conversion subscript candidate from a plurality of types of bases related to the one-dimensional DCT. The subscript that can be used is determined in advance. Next, based on the application result of the application means, a predetermined one that can be determined as a conversion index candidate among a plurality of types of bases related to the one-dimensional DCT, one type of bases related to the two-dimensional DCT, The conversion index candidate was decided. According to this, the types of bases that can be applied to each transform block by the video encoding device are limited. For this reason, since the amount of information about the base transmitted from the video encoding device to the video decoding device can be reduced, the encoding performance can be further improved.

  (5) The present invention is a moving picture decoding apparatus that decodes encoded data generated by the moving picture encoding apparatus according to any one of (1) to (4), wherein the plurality of bases and the plurality of subscripts are used. Decoding side storage means (for example, corresponding to a decoding side storage unit described later) and decoding means for decoding the encoded data and obtaining the conversion index candidate and the encoding information (for example, FIG. 5) Subscript decoding means (for example, FIG. 5) that obtains conversion subscripts based on the entropy decoding unit 101), conversion subscript candidates acquired by the decoding means, and encoded information acquired by the decoding means. The subscript decoding means includes information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied to the encoded information acquired by the decoding means. The conversion subscript candidate acquired by the decoding unit is acquired as the conversion subscript for each conversion block, and only the two-dimensional DCT is applied to the encoded information acquired by the decoding unit. In the case where information is included, for all transform blocks, among the plurality of subscripts stored in the decoding-side storage means, those associated with one type of base related to the two-dimensional DCT The video decoding device is characterized in that it is acquired as the conversion subscript.

  According to the present invention, the moving picture decoding apparatus for decoding the encoded data generated by the moving picture encoding apparatus according to any one of (1) to (4) includes the decoding side storage means, the decoding means, and the subscript decoding means. Was provided. The decoding side storage means stores a plurality of bases and a plurality of subscripts. In addition, the encoded data is decoded by the decoding means to obtain conversion subscript candidates and encoding information. Further, when the subscript decoding means includes information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied to the encoded information acquired by the decoding means, the decoding is performed for each transform block. The conversion index candidate acquired by the means is acquired as a conversion index. In addition, when the subscript decoding means includes information indicating that only the two-dimensional DCT is applied to the encoded information acquired by the decoding means, the decoding side storage means for all transform blocks Among the plurality of subscripts stored in the table, one associated with one type of base related to the two-dimensional DCT is acquired as a conversion subscript.

  For this reason, for each transform block, the same base as the base applied by the moving image coding apparatus can be acquired as a transform subscript and used for decoding the encoded data.

  (6) In the moving picture decoding apparatus according to (5), the decoding unit decodes the encoded data received by the moving picture decoding apparatus after obtaining the encoded information, and converts the converted subscript. A video decoding apparatus characterized by acquiring candidates is proposed.

  According to the present invention, in the moving picture decoding apparatus according to (5), the decoding means obtains the conversion index candidate by decoding the encoded data received by the moving picture decoding apparatus after obtaining the encoded information. did. According to this, an effect similar to the effect mentioned above can be produced.

  (7) The present invention includes an encoding-side storage unit (for example, equivalent to an encoding-side storage unit described later), an application unit (for example, equivalent to the conversion unit 3 and the cost calculation unit 6 in FIG. 1), a determination unit (for example, , Equivalent to the cost calculation unit 6 in FIG. 1), encoding information generation means (for example, equivalent to the cost calculation unit 6 in FIG. 1), and encoding means (for example, equivalent to the entropy encoding unit 5 in FIG. 1). A moving image decoding method in a moving image encoding device that encodes a moving image (e.g., corresponding to an input image to be described later), wherein the encoding side storage means includes a plurality of predetermined bases, A first step of storing a plurality of subscripts provided so that each of the plurality of bases can be identified; and the applying unit applies each of the plurality of bases to each transform block to obtain a transform coefficient Second step and the determination The stage determines an application to be applied among the plurality of bases for each conversion block using the application result obtained by the applying unit, and determines a subscript associated with the determined base for each conversion block as a conversion subscript candidate. A third step, a fourth step in which the encoded information generating means generates encoded information (e.g., corresponding to the slice header of FIG. 1) based on a determination result by the determining means, and the encoding And a fifth step of generating encoded data (for example, corresponding to the bit stream of FIG. 1) based on the determination result by the determining unit, and a plurality of units stored in the encoding-side storage unit Are composed of a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform) and one type of base related to two-dimensional DCT (two-dimensional discrete cosine transform). In the third step, the determining unit determines whether to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner or to apply only the two-dimensional DCT based on the application result of the applying unit. When it is determined that the one-dimensional DCT and the two-dimensional DCT are mixedly applied, an applied one of the plurality of bases is determined for each conversion block, and a subscript associated with the determined base is used as a conversion subscript candidate. If it is determined for each transform block and it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a transform subscript candidate for all transform blocks, and the fourth In the step, if the encoding information generating means determines that the one-dimensional DCT and the two-dimensional DCT are mixed and applied, the determining means determines the one-dimensional D The coding information including information indicating that T and two-dimensional DCT are applied in combination is generated, and when only the two-dimensional DCT is applied, only the two-dimensional DCT is applied when determined by the determining means. In the fifth step, when the encoding unit determines that the encoding unit applies a mixture of one-dimensional DCT and two-dimensional DCT, in the fifth step, In the case where it is determined by the determination means that the encoded data including the conversion index candidate for each conversion block determined by the determination means and the encoding information is generated and only the two-dimensional DCT is applied, A moving image encoding method is proposed, which generates encoded data including the encoding information.

  According to the present invention, the encoding side storage means stores a plurality of predetermined bases and a plurality of subscripts provided so as to be able to identify each of the plurality of bases. The plurality of bases stored in the encoding-side storage means include a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform), a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform), It shall consist of. In addition, the application means determines the transform coefficient by applying it to each transform block for each of the plurality of bases. Further, the decision means decides whether to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner or to apply only the two-dimensional DCT using the application result of the application means. In addition, when the determination unit determines to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner, an applied one of a plurality of bases is determined for each transform block, and a subscript associated with the determined base Is determined for each conversion block as a conversion subscript candidate, and when it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all the conversion blocks. It was decided. In addition, when the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied in a mixed manner by the encoding information generation unit, information indicating that the one-dimensional DCT and the two-dimensional DCT are applied in combination. It was decided to generate the encoded information including. In addition, when only the two-dimensional DCT is applied by the encoding information generation unit, the determination unit determines that the encoding information including information indicating that only the two-dimensional DCT is applied is generated. In addition, when the encoding unit determines that the one-dimensional DCT and the two-dimensional DCT are mixedly applied by the encoding unit, the conversion index candidate for each conversion block determined by the determination unit, the encoding information, When the determination unit determines that only the two-dimensional DCT is applied, the encoded data including the encoding information is generated. For this reason, the effect similar to the effect mentioned above can be produced.

  (8) The present invention provides a decoding-side storage means (for example, equivalent to a later-described decoding-side storage section), a decoding means (for example, equivalent to the entropy decoding section 101 in FIG. 5), and a subscript decoding means (for example, in FIG. A moving picture decoding method in a moving picture decoding apparatus for decoding encoded data generated in the moving picture coding apparatus according to (7), wherein the decoding side storage means A sixth step of storing the plurality of bases and the plurality of subscripts; and a seventh step in which the decoding means decodes the encoded data to obtain the conversion subscript candidates and the encoding information; The subscript decoding means includes an eighth step of obtaining a conversion subscript based on the conversion subscript candidate obtained by the decoding means and the encoded information obtained by the decoding means, and In step 7, when the subscript decoding means includes information indicating that the encoding information acquired by the decoding means is applied in a mixed manner of the one-dimensional DCT and the two-dimensional DCT, conversion is performed. For each block, the conversion index candidate acquired by the decoding unit is acquired as the conversion index, and the encoding information acquired by the decoding unit includes information indicating that only two-dimensional DCT is applied. In this case, for all transform blocks, a plurality of subscripts stored in the decoding-side storage means that are associated with one type of base related to the two-dimensional DCT are obtained as the transform subscripts. A video decoding method characterized by the above is proposed.

  According to this invention, the decoding side storage means stores a plurality of bases and a plurality of subscripts. In addition, the encoded data is decoded by the decoding means to obtain conversion subscript candidates and encoding information. Further, when the subscript decoding means includes information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied to the encoded information acquired by the decoding means, the decoding is performed for each transform block. The conversion index candidate acquired by the means is acquired as a conversion index. In addition, when the subscript decoding means includes information indicating that only the two-dimensional DCT is applied to the encoded information acquired by the decoding means, the decoding side storage means for all transform blocks Among the plurality of subscripts stored in the table, one associated with one type of base related to the two-dimensional DCT is acquired as a conversion subscript. For this reason, the effect similar to the effect mentioned above can be produced.

  (9) The present invention includes an encoding-side storage unit (for example, equivalent to an encoding-side storage unit described later), an application unit (for example, equivalent to the conversion unit 3 and the cost calculation unit 6 in FIG. 1), a determination unit (for example, , Equivalent to the cost calculation unit 6 in FIG. 1), encoding information generation means (for example, equivalent to the cost calculation unit 6 in FIG. 1), and encoding means (for example, equivalent to the entropy encoding unit 5 in FIG. 1). A program for causing a computer to execute a moving image decoding method in a moving image encoding apparatus that encodes a moving image (e.g., corresponding to an input image to be described later). A first step of storing a plurality of defined bases and a plurality of subscripts provided so that each of the plurality of bases can be identified; and the applying unit converts each of the plurality of bases for each conversion Apply to block A second step of obtaining a transform coefficient; and the determination means determines, for each transform block, an application to be applied among the plurality of bases using an application result by the application means, and is associated with the determined base A third step of determining a subscript as a conversion subscript candidate for each conversion block; and the encoding information generation unit, based on the determination result by the determination unit, encode information (for example, corresponding to the slice header of FIG. 1) And a fifth step in which the encoding unit generates encoded data (e.g., corresponding to the bitstream of FIG. 1) based on the determination result by the determination unit. The plurality of bases stored in the encoding-side storage means are a plurality of types of bases related to a one-dimensional DCT (one-dimensional discrete cosine transform) and a two-dimensional DCT ( And in the third step, the determination means mixes the one-dimensional DCT and the two-dimensional DCT based on the application result of the application means. If it is determined whether to apply only two-dimensional DCT, and one-dimensional DCT and two-dimensional DCT are mixedly applied, one of the plurality of bases is applied to each transform block. The subscript associated with the base related to the two-dimensional DCT is converted when the subscript corresponding to the determined base is determined for each conversion block as a conversion subscript candidate, and only the two-dimensional DCT is applied. Subscript candidates are determined for all transform blocks, and in the fourth step, the encoding information generation means mixes one-dimensional DCT and two-dimensional DCT. If it is determined by the determining means, the encoded information including information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied is generated, and the determination is performed when only the two-dimensional DCT is applied. If determined by the means, the encoding information including information indicating that only the two-dimensional DCT is applied is generated, and in the fifth step, the encoding means includes the one-dimensional DCT and the two-dimensional DCT. When the determination means determines that the mixed application is applied, the encoding data including the conversion index candidate for each conversion block determined by the determination means and the encoding information is generated, and two-dimensional If only the DCT is applied and the decision means decides, a program for generating coded data including the coding information is proposed.

  According to this invention, by executing the program, the encoding-side storage unit stores a plurality of predetermined bases and a plurality of subscripts provided so as to be able to identify each of the plurality of bases. It was decided. The plurality of bases stored in the encoding-side storage means include a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform), a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform), It shall consist of. In addition, the application means determines the transform coefficient by applying it to each transform block for each of the plurality of bases. Further, the decision means decides whether to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner or to apply only the two-dimensional DCT using the application result of the application means. In addition, when the determination unit determines to apply the one-dimensional DCT and the two-dimensional DCT in a mixed manner, an applied one of a plurality of bases is determined for each transform block, and a subscript associated with the determined base Is determined for each conversion block as a conversion subscript candidate, and when it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all the conversion blocks. It was decided. In addition, when the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied in a mixed manner by the encoding information generation unit, information indicating that the one-dimensional DCT and the two-dimensional DCT are applied in combination. It was decided to generate the encoded information including. In addition, when only the two-dimensional DCT is applied by the encoding information generation unit, the determination unit determines that the encoding information including information indicating that only the two-dimensional DCT is applied is generated. In addition, when the encoding unit determines that the one-dimensional DCT and the two-dimensional DCT are mixedly applied by the encoding unit, the conversion index candidate for each conversion block determined by the determination unit, the encoding information, When the determination unit determines that only the two-dimensional DCT is applied, the encoded data including the encoding information is generated. For this reason, the effect similar to the effect mentioned above can be produced.

  (10) The present invention provides a decoding-side storage means (for example, equivalent to a later-described decoding-side storage section), a decoding means (for example, equivalent to the entropy decoding section 101 in FIG. 5), and a subscript decoding means (for example, in FIG. A program for causing a computer to execute a video decoding method in the video decoding device that decodes encoded data generated in the video encoding device according to (9). The decoding-side storage means stores the plurality of bases and the plurality of subscripts, and the decoding means decodes the encoded data to convert the conversion subscript candidates and the encoding The seventh step of acquiring information, and the subscript decoding means change based on the conversion subscript candidate acquired by the decoding means and the encoded information acquired by the decoding means. And an eighth step of acquiring a subscript. In the seventh step, the subscript decoding means mixes the one-dimensional DCT and the two-dimensional DCT in the encoded information acquired by the decoding means. If the information indicating application is included, the conversion subscript candidate acquired by the decoding unit is acquired as the conversion subscript for each conversion block, and the encoding information acquired by the decoding unit is acquired. Includes information indicating that only the two-dimensional DCT is applied, among the plurality of subscripts stored in the decoding-side storage unit for all transform blocks, the two-dimensional DCT A program is proposed for acquiring the one associated with one type of base as the conversion subscript.

  According to the present invention, a plurality of bases and a plurality of subscripts are stored in the decoding-side storage unit by executing the program. In addition, the encoded data is decoded by the decoding means to obtain conversion subscript candidates and encoding information. Further, when the subscript decoding means includes information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied to the encoded information acquired by the decoding means, the decoding is performed for each transform block. The conversion index candidate acquired by the means is acquired as a conversion index. In addition, when the subscript decoding means includes information indicating that only the two-dimensional DCT is applied to the encoded information acquired by the decoding means, the decoding side storage means for all transform blocks Among the plurality of subscripts stored in the table, one associated with one type of base related to the two-dimensional DCT is acquired as a conversion subscript. For this reason, the effect similar to the effect mentioned above can be produced.

  ADVANTAGE OF THE INVENTION According to this invention, the increase in the side information transmitted to a moving image decoding apparatus from a moving image encoding apparatus can be suppressed, and encoding performance can be improved.

It is a block diagram which shows the structure of the moving image encoder which concerns on one Embodiment of this invention. It is a figure for demonstrating operation | movement of the conversion part with which the said moving image encoder is provided. It is a figure for demonstrating operation | movement of the inverse transformation part with which the said moving image encoder is provided. It is a figure which shows a part of slice header output from the said moving image encoder. It is a block diagram which shows the structure of the moving image decoding apparatus which concerns on one Embodiment of this invention. It is a figure for demonstrating rearrangement of a pixel. It is a figure for demonstrating rearrangement of a pixel.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the constituent elements in the following embodiments can be appropriately replaced with existing constituent elements and the like, and various variations including combinations with other existing constituent elements are possible. Accordingly, the description of the following embodiments does not limit the contents of the invention described in the claims.

[Configuration of Moving Image Encoding Device AA]
FIG. 1 is a block diagram showing a configuration of a moving picture coding apparatus AA according to an embodiment of the present invention. The moving image coding apparatus AA includes a motion compensation unit 1, a subscript candidate unit 2, a conversion unit 3, an inverse conversion unit 4, an entropy coding unit 5, a cost calculation unit 6, and a coding side storage unit (not shown).

  The encoding side storage unit stores a plurality of predetermined bases and a plurality of subscripts provided so as to identify each of the plurality of bases. Here, the plurality of bases are composed of a one-dimensional DCT or a two-dimensional DCT and a permutation matrix. Further, the permutation matrix is determined in advance according to the transform block size in order to rearrange the pixels along the edge in a line as described above with reference to FIG. That is, the permutation matrix corresponds to the edge direction included in the transform block. The one-dimensional DCT is determined according to the transform block size. Since the one-dimensional DCT depends only on the size of the transform block, that is, the number of coefficients, if the permutation matrix is determined, the one-dimensional DCT is also determined accordingly. In addition, since the two-dimensional DCT does not consider the edge direction, there is no permutation matrix and one is determined according to the transform block size.

  An input image and a decoded image are input to the motion compensation unit 1. Here, the decoded image is a decoded frame. The motion compensation unit 1 obtains a motion vector in the transform block based on these inputs, determines a motion compensation reference block using the obtained motion vector, and outputs the determined reference block and motion vector.

  The subscript candidate unit 2 receives a selection signal (described later) output from the cost calculation unit 6. The subscript candidate unit 2 determines an optimum one from a plurality of bases according to a selection signal for each transform block, and outputs a subscript associated with the determined base.

  The subscript output from the subscript candidate unit 2 is input to the conversion unit 3, and the difference between the input image and the reference block output from the motion compensation unit 1 is input. The conversion unit 3 performs processing on the difference between the input image and the reference block using a base corresponding to the conversion block, and outputs a conversion coefficient. A specific operation of the conversion unit 3 will be described below with reference to FIG.

  FIG. 2 shows a case where one type of base related to the two-dimensional DCT and N types of bases related to the one-dimensional DCT (N is an integer satisfying N ≧ 1) are stored in the encoding side storage unit. Yes. First, the conversion unit 3 selects a base associated with the subscript output from the subscript candidate unit 2 from a plurality of bases stored in the encoding-side storage unit. Next, when the selected base is related to the two-dimensional DCT, 2D transformation and quantization are applied to the difference between the input image and the reference block using the selected base, and the transform coefficient is obtained. Output. Further, when the selected base is related to the one-dimensional DCT, a replacement coefficient is applied to the difference between the input image and the reference block by applying replacement, 1D transform, and quantization using the selected base. Output as.

  Returning to FIG. 1, the reverse conversion unit 4 receives the subscript output from the subscript candidate unit 2 and the conversion coefficient output from the conversion unit 3. The inverse transform unit 4 performs processing on the transform coefficient output from the transform unit 3 using a base corresponding to the transform block, and outputs the result. The sum of the output from the inverse transform unit 4 and the reference block output from the motion compensation unit 1 is input to the motion compensation unit 1 and the cost calculation unit 6 as a decoded image. The specific operation of the inverse conversion unit 4 will be described below with reference to FIG.

  FIG. 3 shows a case where one type of base related to the two-dimensional DCT and N types of bases related to the one-dimensional DCT are stored in the encoding-side storage unit, as in FIG. The inverse transform unit 4 first selects a base associated with the subscript output from the subscript candidate unit 2 from among a plurality of bases stored in the encoding side storage unit. Next, when the selected base is related to the two-dimensional DCT, inverse quantization and 2D inverse transform are applied to the transform coefficient output from the transform unit 3 using the selected base. When the selected base is related to the one-dimensional DCT, inverse quantization, 1D inverse transform, and inverse replacement are applied to the transform coefficient output from the transform unit 3 using the selected base. .

  Returning to FIG. 1, the decoded image input to the motion compensation unit 1, the input image, and the bit stream output from the entropy encoding unit 5 are input to the cost calculation unit 6. The cost calculation unit 6 determines an optimum base for each transform block.

  Specifically, the cost calculation unit 6 first calculates the cost value when applied to each transform block for each basis regarding the one-dimensional DCT and the basis regarding the two-dimensional DCT, using Equation (1). Then, for each transform block, a base that minimizes the cost value is obtained, and the minimum cost value of each transform block is added up for the entire frame. According to this, the cost value when 1D conversion and 2D conversion are mixedly applied to the processing frame is obtained.

  In Equation (1), the generated code amount is the length of the bit stream. Λ is a predetermined coefficient. The coding error is a difference between the input image and the decoded image. A decoded image is generated by encoding an input image using a basis for calculating a cost value and then decoding the encoded result.

  In addition, the cost calculation unit 6 calculates a cost value in the case where a basis related to the two-dimensional DCT is applied for each transform block, using Expression (1). Then, the obtained cost values are added up for the entire frame. According to this, the cost value when only 2D conversion is applied to the processing frame is obtained.

  Next, the cost calculation unit 6 includes a cost value when 1D conversion and 2D conversion are mixedly applied to the processing frame, and a cost value when only 2D conversion is applied to the processing frame. The one with the smaller cost value is applied to the processing frame. Specifically, when the cost value when the 1D conversion and the 2D conversion are mixedly applied to the processing frame is smaller, the conversion is performed using the basis that minimizes the cost value for each conversion block. The subscript associated with the determined base is output as the selection signal. On the other hand, when the cost value when only 2D conversion is applied to the processing frame is smaller, it is determined that the processing frame is converted using the basis related to the two-dimensional DCT, and the determined base is used. The associated subscript is output as the above-described selection signal.

  In addition, the cost calculation unit 6 describes a value indicating the conversion type in the SH (slice header) flag “trans_dim_type” illustrated in FIG. 4.

  FIG. 4 is a diagram illustrating a part of a slice header output from the video encoding device AA. The flag “first_mb_in_slice” indicates the position of the first macroblock existing in the slice corresponding to the entire screen. The flag “slice_type” indicates the type of slice, and indicates any of an I slice, a P slice, and a B slice. The flag “pic_parameter_set_id” indicates a number (ID) used in this slice in the PPS (Picture Parameter Set). The flag “slice_qp_delta” indicates the difference between the current quantization step and the quantization step in this slice for the quantization step applied to the macroblock in the slice.

  The flag “trans_dim_type” indicates a conversion type, that is, whether 1D conversion and 2D conversion are mixedly applied or only 2D conversion is applied. Specifically, when the flag “trans_dim_type” is “0”, it indicates that only 2D conversion is applied. On the other hand, when the flag “trans_dim_type” is “1”, it indicates that 1D conversion and 2D conversion are applied together.

  Returning to FIG. 1, the entropy encoding unit 5 converts the conversion coefficient output from the conversion unit 3, the conversion subscript output from the subscript candidate unit 2, the motion vector output from the motion compensation unit 1, and the conversion A slice header in which a value indicating the type is described is input. The entropy encoding unit 5 encodes the input signal and outputs a bit stream.

  However, the entropy encoding unit 5 does not include the conversion subscript output from the subscript candidate unit 2 in the bitstream depending on the value indicating the conversion type described in the slice header. Specifically, when the flag “trans_dim_type” of the slice header is “0”, the entropy encoding unit 5 excludes the conversion subscript from the signal to be encoded and outputs a bitstream.

[Configuration of Video Decoding Device BB]
FIG. 5 is a block diagram showing a configuration of a video decoding device BB according to an embodiment of the present invention. The video decoding device BB includes an entropy decoding unit 101, an inverse transformation unit 102, a subscript decoding unit 103, a motion compensation unit 104, and a decoding side storage unit (not shown), and bits generated by the video encoding device AA. Decode the stream.

  The decoding-side storage unit stores a plurality of predetermined bases and a plurality of subscripts provided so as to identify each of the plurality of bases. The plurality of bases and the plurality of subscripts stored in the decoding-side storage unit are the same as the plurality of bases and the plurality of subscripts stored in the above-described encoding-side storage unit.

  The entropy decoding unit 101 receives a bit stream. The entropy decoding unit 101 decodes the input bitstream, acquires transform coefficients, transform index candidates, slice headers, and motion vectors, and outputs them. The entropy decoding unit 101 acquires the transform coefficient and the conversion index candidate by decoding the bitstream received by the video decoding device BB after acquiring the slice header.

  The subscript decoding unit 103 receives the conversion subscript candidate and the slice header output from the entropy decoding unit 101. When the conversion type included in the slice header is a single type of base, that is, when the cost calculation unit 6 determines that only the 2D conversion is applied, the subscript decoding unit 103 uses the base related to the two-dimensional DCT. The subscript associated with is output as a conversion subscript regardless of the conversion subscript candidate. On the other hand, when the conversion types included in the slice header are a plurality of types of bases, that is, when the cost calculation unit 6 determines to apply 1D conversion and 2D conversion in a mixed manner, conversion subscript candidates are converted into conversion subscripts. Output as.

  The inverse transform unit 102 receives the transform coefficient output from the entropy decoding unit 101 and the transform subscript output from the subscript decoding unit 103. The inverse conversion unit 102 first selects a base associated with the conversion subscript from a plurality of bases stored in the decoding side storage unit. Next, if the selected base is related to the one-dimensional DCT, the selected base is used to apply inverse quantization, 1D inverse transform, and inverse permutation to the transform coefficient, and to obtain a motion compensation residual. Output the difference. If the selected base is related to the two-dimensional DCT, the selected base is used to apply inverse quantization and 2D inverse transform to the transform coefficient to output a motion compensation residual. Then, a sum of a motion compensation residual and a motion compensation signal (described later) output from the motion compensation unit 104 is output as a decoded image from the video decoding device BB.

  The motion compensation unit 104 receives the motion vector output from the entropy decoding unit 101 and the decoded image. The motion compensation unit 104 applies motion compensation to the decoded image using a motion vector, and outputs a motion compensation signal.

  According to the above moving picture coding apparatus AA, the base to be applied is determined for each transform block. Then, when the basis regarding the one-dimensional DCT is determined in at least a part of the transform blocks in the frame, the transform index candidate associated with the basis determined for each transform block, the moving image, and the determined basis are A bit stream is generated by encoding a slice header including information indicating that each transform block is different. On the other hand, when bases related to the two-dimensional DCT are determined in all transform blocks in the frame, a moving image and a slice header including information indicating that the determined base is the same in all transform blocks in the frame; Are encoded to generate a bitstream. For this reason, when bases relating to the two-dimensional DCT are determined in all transform blocks in the frame, transform subscript candidates are not included in the bitstream. Therefore, an increase in side information transmitted from the moving image encoding device AA to the moving image decoding device BB can be suppressed, and the encoding performance can be improved.

  Further, according to the moving picture coding apparatus AA, for each transform block, a generated code amount when each of a plurality of bases is applied is obtained, and the obtained generated code amount is used as shown in Equation (1). Thus, a conversion index candidate is determined for each conversion block. For this reason, the base to be applied can be determined for each transform block so that the generated code amount for each transform block is reduced. Therefore, the encoding performance can be further improved.

  Also, according to the moving image encoding apparatus AA, for each transform block, each of a plurality of bases is applied and encoded, and then decoded to generate a decoded image. As shown in Equation (1), decoding is performed. Using the difference between the image and the input image, a conversion index candidate is determined for each conversion block. For this reason, the base to be applied can be determined for each transform block so that the degradation of the decoded image with respect to the input image for each transform block is suppressed. Therefore, the encoding performance can be further improved.

  Also, according to the video decoding device BB, the bit stream is decoded to obtain the slice header and conversion subscript candidates. When the conversion type included in the slice header is a single type of base, that is, when it is determined that only the 2D conversion is applied in the cost calculation unit 6 described above, the base is associated with the base related to the two-dimensional DCT. Subscripts are converted into subscripts regardless of conversion subscript candidates, and are applied to each conversion block. On the other hand, when the conversion types included in the slice header are a plurality of types of bases, that is, when the cost calculation unit 6 determines to apply 1D conversion and 2D conversion in a mixed manner, conversion subscript candidates are converted into conversion subscripts. And applied to each transform block. For this reason, the encoded data can be decoded by applying the same base as the base applied by the video encoding device AA for each transform block.

  Note that the processing of the moving image encoding device AA and the processing of the moving image decoding device BB of the present invention are stored in a computer-readable recording medium, and the program recorded on the recording medium is stored in the moving image encoding device AA or the moving image. The present invention can be realized by causing the decryption apparatus BB to read and execute it.

  In addition, the above-described program is transferred from the video encoding device AA or the video decoding device BB storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. May be transmitted. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.

  Further, the above-described program may be for realizing a part of the above-described function. Furthermore, what can implement | achieve the above-mentioned function in combination with the program already recorded on the moving image encoder AA and the moving image decoder BB, what is called a difference file (difference program) may be sufficient.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes a design that does not depart from the gist of the present invention.

  For example, in the above-described embodiment, information indicating whether 1D conversion and 2D conversion are mixedly applied or only 2D conversion is applied, that is, the conversion type is described in the slice header. For example, it may be described in a picture parameter set.

  In the above-described embodiment, the moving image encoding device AA and the moving image decoding device BB apply the same base to the same transform block. For this reason, the video decoding device BB needs to be able to recognize all bases that can be applied to each transform block in the video encoding device AA. Therefore, the video encoding device AA needs to transmit in advance information about all bases that can be applied to each transform block to the video decoding device BB. Therefore, for example, a subscript that can be output as a selection signal by the cost calculation unit 6 is determined in advance from a plurality of subscripts stored in the encoding side storage unit, and the cost calculation unit 6 Of the plurality of subscripts stored in the storage unit, the subscript output as the selection signal may be determined from those determined in advance as being output as the selection signal. According to this, the types of bases that can be applied to each transform block by the video encoding device AA are limited. Therefore, the amount of information about the base transmitted from the video encoding device AA to the video decoding device BB can be reduced, so that the encoding performance can be further improved.

DESCRIPTION OF SYMBOLS 1 ... Motion compensation part 2 ... Subscript candidate part 3 ... Conversion part 4 ... Inverse conversion part 5 ... Entropy encoding part 6 ... Cost calculation part 101 ... Entropy decoding part 102 ... Inverse conversion unit 103 ... Subscript decoding unit 104 ... Motion compensation unit AA ... Moving image encoding device BB ... Moving image decoding device

Claims (10)

A moving image encoding device for encoding a moving image,
Encoding side storage means for storing a plurality of predetermined bases and a plurality of subscripts provided so as to identify each of the plurality of bases;
For each of the plurality of bases, application means for applying a transform coefficient to each transform block;
A determination unit that determines, for each conversion block, one to be applied among the plurality of bases, using a result of application by the application unit, and determines a subscript associated with the determined base for each conversion block as a conversion subscript candidate; ,
Encoding information generating means for generating encoding information based on the determination result by the determining means;
Encoding means for generating encoded data based on the determination result by the determining means,
The plurality of bases stored in the encoding-side storage means are a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform) and a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform). Configured,
The determining means is based on the application result by the applying means,
Decide whether to apply a mixture of one-dimensional DCT and two-dimensional DCT or only two-dimensional DCT,
When it is determined that the one-dimensional DCT and the two-dimensional DCT are mixedly applied, an applied one of the plurality of bases is determined for each conversion block, and a subscript associated with the determined base is used as a conversion subscript candidate. Decide for each transformation block,
When it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all conversion blocks,
The encoded information generating means includes
When the one-dimensional DCT and the two-dimensional DCT are mixedly applied and determined by the determining means, the encoding information including information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied is generated. ,
If it is determined by the determining means that only two-dimensional DCT is applied, the encoding information including information indicating that only two-dimensional DCT is applied is generated,
The encoding means includes
If the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied together, the encoding includes the conversion index candidate for each conversion block determined by the determination unit and the encoding information Generate data,
A moving picture coding apparatus that generates coded data including the coding information when only the two-dimensional DCT is applied by the decision means. The moving image encoding device according to claim 1,
The determination unit obtains a code amount when each of the plurality of bases is applied for each transform block, and determines the transform index candidate using the obtained code amount. . The moving picture encoding apparatus according to claim 1 or 2,
The determining means generates a decoded image by encoding after applying each of the plurality of bases for each conversion block, and using the difference between the decoded image and the moving image, the conversion index candidate A video encoding device characterized by determining In the moving image encoder according to any one of claims 1 to 3,
The determining means includes
Of a plurality of types of bases related to the one-dimensional DCT, subscripts that can be determined as the conversion subscript candidates are determined in advance,
Based on the result of application by the application means, a predetermined one of a plurality of types of bases related to the one-dimensional DCT can be determined as the conversion index candidate, one type of bases related to the two-dimensional DCT, The conversion encoding candidate is determined from the video encoding device. A video decoding device for decoding encoded data generated in the video encoding device according to any one of claims 1 to 4,
Decoding side storage means for storing the plurality of bases and the plurality of subscripts;
Decoding means for decoding the encoded data to obtain the conversion index candidates and the encoding information;
A subscript candidate obtained by the decoding means, and subscript decoding means for obtaining a conversion subscript based on the encoding information obtained by the decoding means,
The subscript decoding means is:
When the encoding information acquired by the decoding means includes information indicating that one-dimensional DCT and two-dimensional DCT are mixedly applied, the information is acquired by the decoding means for each transform block Obtain conversion subscript candidates as the conversion subscript,
When the encoding information acquired by the decoding means includes information indicating that only the two-dimensional DCT is applied, it is stored in the decoding storage means for all transform blocks. A moving picture decoding apparatus, wherein a plurality of subscripts associated with one type of base related to the two-dimensional DCT are acquired as the conversion subscript. The moving picture decoding apparatus according to claim 5,
The moving picture decoding apparatus characterized in that the decoding means acquires the conversion subscript candidates by decoding the encoded data received by the moving picture decoding apparatus after acquiring the encoded information. A moving picture decoding method in a moving picture coding apparatus for coding a moving picture, comprising coding side storage means, applying means, determining means, coding information generating means, and coding means,
A first step in which the encoding-side storage unit stores a plurality of predetermined bases and a plurality of subscripts provided so as to be able to identify each of the plurality of bases;
A second step in which the applying means applies a transform coefficient to each transform block for each of the plurality of bases;
The determining means determines, for each conversion block, the one to be applied among the plurality of bases using the application result obtained by the applying means, and uses the subscript associated with the determined base as a conversion subscript candidate for each conversion block. A third step of determining;
A fourth step in which the encoded information generating means generates encoded information based on a determination result by the determining means;
The encoding unit includes a fifth step of generating encoded data based on a determination result by the determination unit;
The plurality of bases stored in the encoding-side storage means are a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform) and a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform). Configured,
In the third step, the determining means is based on an application result by the applying means,
Decide whether to apply a mixture of one-dimensional DCT and two-dimensional DCT or only two-dimensional DCT,
When it is determined that the one-dimensional DCT and the two-dimensional DCT are mixedly applied, an applied one of the plurality of bases is determined for each conversion block, and a subscript associated with the determined base is used as a conversion subscript candidate. Decide for each transformation block,
When it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all conversion blocks,
In the fourth step, the encoded information generating means is
When the one-dimensional DCT and the two-dimensional DCT are mixedly applied and determined by the determining means, the encoding information including information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied is generated. ,
If it is determined by the determining means that only two-dimensional DCT is applied, the encoding information including information indicating that only two-dimensional DCT is applied is generated,
In the fifth step, the encoding means includes
If the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied together, the encoding includes the conversion index candidate for each conversion block determined by the determination unit and the encoding information Generate data,
A moving picture coding method characterized in that, when only the two-dimensional DCT is applied, when the decision means decides, coded data including the coding information is generated. A moving picture decoding method in a moving picture decoding apparatus, comprising decoding side storage means, decoding means, and subscript decoding means, for decoding encoded data generated in the moving picture encoding apparatus according to claim 7,
A sixth step in which the decoding side storage means stores the plurality of bases and the plurality of subscripts;
A seventh step in which the decoding means decodes the encoded data to obtain the conversion index candidate and the encoded information;
The subscript decoding means comprises an eighth step of acquiring a conversion subscript based on the conversion subscript candidate acquired by the decoding means and the encoded information acquired by the decoding means;
In the seventh step, the subscript decoding means is
When the encoding information acquired by the decoding means includes information indicating that one-dimensional DCT and two-dimensional DCT are mixedly applied, the information is acquired by the decoding means for each transform block Obtain conversion subscript candidates as the conversion subscript,
When the encoding information acquired by the decoding means includes information indicating that only the two-dimensional DCT is applied, it is stored in the decoding storage means for all transform blocks. A moving picture decoding method, wherein a plurality of subscripts associated with one type of base relating to the two-dimensional DCT are acquired as the conversion subscript. A program for causing a computer to execute a moving image decoding method in a moving image encoding apparatus that includes an encoding side storage unit, an applying unit, a determining unit, an encoding information generation unit, and an encoding unit and encodes a moving image Because
A first step in which the encoding-side storage unit stores a plurality of predetermined bases and a plurality of subscripts provided so as to be able to identify each of the plurality of bases;
A second step in which the applying means applies a transform coefficient to each transform block for each of the plurality of bases;
The determining means determines, for each conversion block, the one to be applied among the plurality of bases using the application result obtained by the applying means, and uses the subscript associated with the determined base as a conversion subscript candidate for each conversion block. A third step of determining;
A fourth step in which the encoded information generating means generates encoded information based on a determination result by the determining means;
The encoding means causes the computer to execute a fifth step of generating encoded data based on the determination result by the determining means,
The plurality of bases stored in the encoding-side storage means are a plurality of types of bases related to one-dimensional DCT (one-dimensional discrete cosine transform) and a single type of base related to two-dimensional DCT (two-dimensional discrete cosine transform). Configured,
In the third step, the determining means is based on an application result by the applying means,
Decide whether to apply a mixture of one-dimensional DCT and two-dimensional DCT or only two-dimensional DCT,
When it is determined that the one-dimensional DCT and the two-dimensional DCT are mixedly applied, an applied one of the plurality of bases is determined for each conversion block, and a subscript associated with the determined base is used as a conversion subscript candidate. Decide for each transformation block,
When it is determined that only the two-dimensional DCT is applied, the subscript associated with the base relating to the two-dimensional DCT is determined as a conversion subscript candidate for all conversion blocks,
In the fourth step, the encoded information generating means is
When the one-dimensional DCT and the two-dimensional DCT are mixedly applied and determined by the determining means, the encoding information including information indicating that the one-dimensional DCT and the two-dimensional DCT are mixedly applied is generated. ,
If it is determined by the determining means that only two-dimensional DCT is applied, the encoding information including information indicating that only two-dimensional DCT is applied is generated,
In the fifth step, the encoding means includes
If the determination unit determines that the one-dimensional DCT and the two-dimensional DCT are applied together, the encoding includes the conversion index candidate for each conversion block determined by the determination unit and the encoding information Generate data,
A program for generating encoded data including the encoding information when the determination unit determines that only two-dimensional DCT is applied. 10. A moving picture decoding method in a moving picture decoding apparatus for decoding encoded data generated in the moving picture encoding apparatus according to claim 9, comprising a decoding side storage means, a decoding means, and a subscript decoding means. A program for
A sixth step in which the decoding side storage means stores the plurality of bases and the plurality of subscripts;
A seventh step in which the decoding means decodes the encoded data to obtain the conversion index candidate and the encoded information;
The subscript decoding means causes the computer to execute a conversion subscript candidate acquired by the decoding means and an eighth step of acquiring a conversion subscript based on the encoding information acquired by the decoding means;
In the seventh step, the subscript decoding means is
When the encoding information acquired by the decoding means includes information indicating that one-dimensional DCT and two-dimensional DCT are mixedly applied, the information is acquired by the decoding means for each transform block Obtain conversion subscript candidates as the conversion subscript,
When the encoding information acquired by the decoding means includes information indicating that only the two-dimensional DCT is applied, it is stored in the decoding storage means for all transform blocks. A program for acquiring, as a conversion subscript, a plurality of subscripts associated with one type of base related to the two-dimensional DCT.

Images