JP2007513566A - Video coding method and apparatus - Google Patents

Abstract

The present invention provides a method and apparatus for quickly determining an encoding mode during moving image encoding and encoding a moving image under the encoding mode.
In the video encoding method according to the present invention, (a) among a plurality of encoding modes used for video encoding, a block having a predetermined size constituting a video satisfies a condition corresponding to a predetermined first mode. A step of determining whether or not the condition is satisfied; and (b) a step of determining the first mode as a block coding mode when a predetermined condition is satisfied. As a result, it is possible to quickly encode a video without causing a significant deterioration in image quality.

Description

  The present invention relates to moving image encoding, and more particularly, to a method and apparatus for quickly determining an encoding mode during moving image encoding and encoding a moving image under the encoding mode.

  H. on video compression In order to encode a moving picture according to the H.264 standard, one picture is divided into macro blocks. Then, after encoding each macroblock in all encoding modes in interpretation and in all encoding modes in intraprediction, the bit rate required for encoding the macroblock and the original One encoding mode is determined according to the degree of distortion between the macroblock and the decoded macroblock.

  The inter mode encodes a difference between a pixel value and a motion vector information indicating a position of one or a plurality of blocks selected from a reference picture in order to encode a macroblock of a current picture. Is the mode used in H. In the H.264 standard, since there can be up to five reference pictures, the reference picture is searched in the frame memory in which the reference picture is stored, and the block to which the current macroblock refers is searched. The reference picture stored in the frame memory can be a past picture or a future picture based on the current picture.

  Intra mode does not refer to a reference picture to encode a macroblock of the current picture, but uses a pixel value that is spatially adjacent to the macroblock to be encoded and uses the macro value to be encoded. This mode is used in intra-prediction in which the difference between the predicted value and the pixel value is encoded after calculating the predicted value for the block.

  However, in the inter mode, there are a large number of modes depending on how the macroblock is divided, and in the intra mode, there are various modes depending on the prediction direction. Therefore, determining the optimum mode among all such modes is important for determining video coding performance. For this reason, a rate-distortion cost (RDcost) is calculated for all possible modes, and the mode having the smallest value is selected as the encoding mode. However, this process has a problem in that it takes much time and cost to encode the video.

  Therefore, the technical problem to be solved by the present invention is to determine a coding mode in accordance with a predetermined order in coding mode determination, thereby quickly determining a coding mode and coding a moving image. A method and apparatus is provided.

  The technical problem is that according to the present invention, in the video encoding method, (a) among a plurality of encoding modes used for encoding the video, a block of a predetermined size constituting the video is predetermined. Determining whether or not a condition corresponding to the first mode is satisfied; and (b) determining that the first mode is an encoding mode of the block if the predetermined condition is satisfied; It is achieved by an encoding method characterized by including:

  Further, the technical problem is that, in the video encoding method, (a) temporal similarity between a block of a predetermined size in the current picture constituting the video and a reference block of a reference picture under the inter mode And (b) omitting the encoding process under the intra mode according to the calculated temporal similarity and the difference in spatial similarity between the block and the block adjacent to the current picture. Determining whether or not to perform the encoding method.

  The technical problem is also achieved by the above-described apparatus for video encoding.

  The technical problem is also achieved by a computer-readable medium in which a program for performing the above-described video encoding method is implemented.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

  The encoding device according to the present invention encodes moving image data. The moving image data is composed of a plurality of pictures arranged on the time axis, and the picture is composed of a plurality of blocks. The block includes a macroblock and a subblock obtained by dividing the macroblock into two or four in the vertical or horizontal direction. The division of the macroblock will be described later with reference to FIG.

  FIG. 1 is a block diagram of an encoding apparatus according to a preferred embodiment of the present invention.

  The encoding apparatus includes a motion estimation unit 102, a motion compensation unit 104, an intra-prediction execution unit 106, a conversion unit 108, a quantization unit 110, a rearrangement unit 112, an entropy coding unit 114, an inverse quantization unit 116, an inverse variation. A return unit 118, a filter 120, and a frame memory 122 are provided.

  The encoding apparatus performs encoding on the macroblock of the current picture under one encoding mode selected from various encoding modes. For this purpose, encoding is performed under all modes that the above-described interpretation and intra-prediction may have, RDcost is calculated, and the mode having the smallest value is determined as the optimum mode, Encoding is performed under that mode. RDcost will be described later. However, the optimum mode can be determined by the encoding mode determination method of the present invention by various methods in addition to the calculation of RDcost.

  The motion estimation unit 102 searches the reference picture for the predicted value of the macroblock of the current picture for interpretation. Then, when a reference block is found in units of 1/2 pixel or 1/4 pixel, the motion compensation unit 104 determines the reference block data value by calculating these intermediate pixel values. In this way, the interpretation is performed by the motion estimation unit 102 and the motion compensation unit 104.

  Also, intra prediction execution unit 106 performs intra prediction for searching for a predicted value of a macroblock of the current picture in the current picture. Whether the current macroblock is to be interpreted or intrapredicted is calculated by calculating the RDcost under all coding modes, and selecting the mode with the smallest value as the code of the block. Encoding is performed on the macroblock.

  As described above, when the prediction data referred to by the macroblock of the current frame is found through the execution of the interpredation or intraprediction, the prediction data is subtracted from the macroblock of the current picture, and converted by the conversion unit 108. After performing the transformation at, the quantization unit 110 performs the quantization. A difference obtained by subtracting a motion-estimated reference block from a macroblock of the current frame is called a residual, which is to encode a residual value in order to reduce the amount of data at the time of encoding. The quantized residual value passes through the rearrangement unit 112 for encoding by the entropy coding unit 114.

  On the other hand, in order to obtain a reference picture used for interpretation, the current picture is restored by passing the quantized picture through the inverse quantization unit 116 and the inverse transform unit 118. The current picture restored in this way is stored in the frame memory and used for execution of interpretation for the next picture. If the restored picture passes through the filter 120, the original picture becomes a picture including some coding errors.

  FIG. 2 is a diagram illustrating variable blocks that a macroblock may have at the time of interpretation.

  H. In the H.264 interpretation, one 16 × 16 macroblock is divided into 16 × 16, 16 × 8, 8 × 16, or 8 × 8 blocks, and each 8 × 8 block is divided into smaller units. 8 × 4, 4 × 8, and 4 × 4 blocks. Motion estimation and compensation are performed on each of the sub-blocks thus divided to determine a motion vector. If interpretation is performed using such various types of variable blocks, encoding can be effectively performed according to video characteristics and motion.

  FIG. 3 is a diagram illustrating an example of a block shape allocated to each macroblock of a current picture to be encoded. As shown in FIG. 3, when the video motion is small and the object is large, a large block is used. When the video motion is complex and the object is small, the small block is used to interpret. It can be seen that the size of the variable block is assigned to perform the operation.

  Therefore, there are many types of subblocks that one macroblock can have. As described above, one macro block is divided into 16 × 8, 8 × 16, and 8 × 8, and each 8 × 8 block is further divided into 8 × 4, 4 × 8, and 4 × 4 blocks. As a result, there are four types of modes that a single macroblock can have: 16 × 16, 16 × 8, 8 × 16, and 8 × 8, and there are four types of each 8 × 8 block. Therefore, there are (4 × 4 × 4 × 4-1) = 255 types, and all 4 + 255 = 259 types of modes exist. Therefore, in the interpretation, since RDcost must be calculated for all 259 types of block shapes and one mode must be determined, the amount of calculation is very large.

  FIG. 4 is a diagram illustrating a luminance (luma) block P to be predicted and adjacent blocks used for prediction.

  Intraprediction is performed for luminance and chromaticity, respectively. If a block is encoded in the intra mode, the block P410 to be predicted is predicted using the decoded blocks A to L adjacent to the block. The prediction is performed not only on the luminance block but also on the chromaticity (U signal, V signal) block. The luminance prediction block P410 is a 16 × 16 block composed of a plurality of 4 × 4 blocks. Here, a to p are 4 × 4 blocks to be predicted, and A, B, C, D and I, J, K, and L are blocks used for prediction.

  There are two types of intra prediction, 4 × 4 prediction and 16 × 16 prediction, depending on the size of the block to be predicted. Depending on the direction of prediction, there are 9 types of modes for 4x4 prediction and 4 types of modes for 16x16 prediction. The 4 × 4 prediction mode is divided according to a prediction direction when a prediction sample is obtained using pixel values of blocks A, B, C, D and I, J, K, and L adjacent to the 4 × 4 block to be predicted.

  More specifically, the 4 × 4 intra luminance prediction mode includes a vertical mode, a horizontal mode, a DC mode, a diagonal_down_left mode, a diagonal_down_right mode, a vertical_right mode, a horizontal_down mode, and a vertical mode, vertical mode, and vertical mode. There are four types of 16 × 16 intra luminance prediction modes: a vertical mode, a horizontal mode, a DC mode, and a plane mode. Therefore, since the RDcost in these various modes must be calculated, the amount of calculation is very large as in the case of interpretation.

  5A to 5B are diagrams for explaining calculation of a predicted motion vector (PMV).

  As will be described with reference to FIG. 5A, the current macroblock C has a high probability of being similar to the motion vector of the adjacent macroblock. Therefore, the motion vector of the current macroblock C is estimated using the motion vectors of A, B, and D, which are macroblocks adjacent to C. As an example of estimation, an intermediate value of motion vectors of A, B, and D macroblocks can be taken. The motion vector estimated in this way is set as PMV, and at the time of encoding, the PMV is subtracted from the actual motion vector for encoding.

  As shown in FIG. 5B, if the size of the block of the current block C ′ is different from the size of the divided block of the adjacent macroblock, any of the divided blocks of the adjacent macroblock is changed to the adjacent block. It becomes a problem whether to decide as. At this time, in the block adjacent to the upper side of the current block, the A ′ block that is the leftmost block is selected, and the rightmost pixel on the upper side of the current block and the block B ′ that is adjacent to the upper right side diagonally. In the block and the block adjacent to the left side of the current block, the D ′ block which is the uppermost block is selected. The PMV is calculated using the motion vectors of the blocks A ′, B ′, and D ′ thus selected.

  H. At the time of H.264 encoding, prediction is performed by selecting any one mode from one mode selected from the above-described interpretation and one mode selected from intra-prediction. Depending on what mode is selected for prediction, H. H.264 compression efficiency changes, but in order to select the best mode, block prediction is performed by applying all modes, and RDcost is calculated.

  FIG. 6 is a diagram for explaining calculation of RDcost.

  Rate (R) means a bit rate and represents the number of bits used for encoding one macroblock. Therefore, the rate is the number of bits obtained by passing the residual signal after the execution of the interprediction or intraprediction through the transform and quantization unit 610 and the variable length coding unit 620, and the motion. This is a value obtained by adding together the number of bits when the vector information passes through the variable length coding unit 620 and the number of bits when the reference picture information passes through the variable length coding unit 620. On the other hand, distortion (D) is the difference between the original macroblock when the encoded video is decoded and the decoded macroblock. Therefore, the distortion D is a value that can be understood if the original macroblock is restored through the inverse quantization and inverse transform unit 630. The RDcost is calculated by the following equation (1) by the RDcost calculation unit 640 having the rate and distortion obtained as described above.

式（１）でＤｉｓｔｏｒｔｉｏｎは、現在マクロブロックとエンコーディングされて、デコーディングされて復元された該マクロブロックとの画素値の差を意味し、次の式（２）によって計算される。そして、Ｒａｔｅは、前述した方法により計算されたビット率を意味する。

In Expression (1), “Distortion” means a difference in pixel value between the current macroblock and the decoded macroblock, and is calculated by the following Expression (2). “Rate” means a bit rate calculated by the above-described method.

式（２）で、Ｂ（ｋ，ｌ）及びＢ’（ｋ，ｌ）は、それぞれ現在マクロブロック及び復元されたマクロブロックの（ｋ，ｌ）番目の画素値を表す。λの一例は、次の式（３）のような式によって決定され得る。

In Equation (2), B (k, l) and B ′ (k, l) represent the (k, l) -th pixel values of the current macroblock and the restored macroblock, respectively. An example of λ can be determined by the following equation (3).

式（３）で、ＱＰは、０ないし５１の整数であって、Ｈ．２６４量子化値である。Ｒａｔｅは、現在マクロブロックの符号化に所要されたビット数である。

In the formula (3), QP is an integer of 0 to 51, and H.P. H.264 quantization value. Rate is the number of bits required for encoding the current macroblock.

  FIG. 7 is a flowchart of a method for determining an encoding mode.

  If a current block is input, it is preferentially checked whether the block can be determined as a skip mode. The skip mode is a mode in which only the encoding mode information of the block to be encoded is transmitted or stored. In the background part of the video, etc., there is a high probability that the block given the current picture has the same pixel value as the corresponding block of the reference picture. In this case, a separate signal such as a residual signal or motion vector information is used. This is because only the encoding mode information needs to be transmitted or stored without the need to transmit or store the encoded data. In order to check whether the skip mode can be determined, motion estimation is performed on the current block, and a reference picture is determined (S705). Then, it is determined whether or not the current block satisfies the skip mode condition (S710). In order to be determined as the skip mode, all of the following four cases must be satisfied.

  (1) The size of the block used for interpretation is 16 × 16, and (2) the reference picture used for encoding the current block is the current picture and time among the reference pictures stored in the frame memory. (3) the motion vector is (0, 0) or the same as PMV, and (4) the residual that is the difference in pixel values between the current block and the reference picture block This is a case where the value after the value is converted and quantized is 0.

  Or you must satisfy all the following three cases. (1) The size of a block used for interpretation is 8x8, and (2) a reference picture and motion vector used for encoding the current block are spatially or temporally adjacent blocks of the current block. (3) The value after converting and quantizing the residual value, which is the difference between the pixel values of the current block and the block of the reference picture, is 0. Is the case.

  If all the skip conditions described above are satisfied, the block is determined as a skip mode, and the process ends without checking the possibility of entering another mode (S720).

  If the skip condition described above is not satisfied, the interpredation is performed under other modes that may be included in the interpretation, and one of the intermodes is determined, and then encoding is performed under the determined intermode. The number of bits when performing is calculated (S725). At this time, instead of targeting all possible modes, any one inter mode among groups including some modes may be determined as necessary. In this way, an average rate (AR) is calculated. The average rate AR is calculated by the following equation (4) and represents the rate per pixel. Since Interpretation has already been performed, Rate is a value already known.

そして、平均境界エラー（Ａｖｅｒａｇｅ Ｂｏｕｎｄａｒｙ Ｅｒｒｏｒ；ＡＢＥ）を計算する（Ｓ７３０）。

Then, an average boundary error (ABE) is calculated (S730).

  8A to 8B are diagrams for explaining the calculation of the average boundary error ABE.

  In order to calculate the average boundary error ABE, the sum of boundary errors (Sum of Boundary Error: SBE) is calculated by the following equation (5). Thereafter, an average boundary error ABE per pixel is calculated.

ここで、Ｙ_ｏｒｇ、Ｕ_ｏｒｇ、Ｖ_ｏｒｇは、符号化される現在マクロブロックの画素値を意味し、Ｙ_Ｒｅｃ、Ｕ_Ｒｅｃ、Ｖ_Ｒｅｃは、復号化された隣接マクロブロックの画素値を意味する。図８Ａに示すように、また（ｘ，ｙ）は、輝度ブロックの画素座標値を、（ｃｘ，ｃｙ）は、色度ブロックの画素座標値を表す。

Here, Y _org , U _org , and V _org mean the pixel values of the current macroblock to be encoded, and Y _Rec , U _Rec , and V _Rec mean the pixel values of the decoded neighboring macroblock. . As shown in FIG. 8A, (x, y) represents the pixel coordinate value of the luminance block, and (cx, cy) represents the pixel coordinate value of the chromaticity block.

  In the first case shown in FIG. 8B, since there is no decoded adjacent macroblock, the average boundary error ABE cannot be calculated by the above-described equation (5). Therefore, in this case, ABE = 0 is determined. When there is no left adjacent block as in the second case, the above-described equation (5) is as the following equation (6).

第３の場合のように、上側の隣接ブロックが存在しない場合には、前述した式（５）は、次の式（７）の通りである。

When there is no upper adjacent block as in the third case, the above-described equation (5) is as the following equation (7).

第４の場合は、既に説明した式（５）によって平均境界エラーを求めうる。

In the fourth case, the average boundary error can be obtained by Equation (5) already described.

  The average rate AR calculated in this way is compared with the average boundary error ABE (S735). Average rate and average boundary error ABE are examples of calculating temporal similarity and spatial similarity, respectively. If the average boundary error ABE is larger than the average rate AR, it means that the probability that the inter mode is determined is higher than the intra mode. Therefore, the macro block is determined to be the inter mode determined in S725, and the process ends (S750). If the average boundary error ABE is not greater than the average rate AR, one of the intra modes is selected (S740). Finally, one of the determined intra mode and the already determined inter mode is selected to determine the macroblock coding mode (S745).

  Yet another example of determining temporal similarity is to use the difference in pixel values between the macroblock in the current picture and the reference block in the reference picture under inter mode. This difference is calculated by calculating the difference between each pixel value constituting the current macroblock and the pixel value in the reference block of the reference picture corresponding to each pixel position, and then obtaining the sum or square sum of those absolute values. Later, it can be obtained by multiplying by a predetermined ratio λ.

  FIG. 9 is a simplified block diagram of the encoding apparatus of the present invention.

  The encoding device includes a determination unit 910, an encoding mode determination unit 920, and an encoding execution unit 930.

  The determination unit 910 determines whether or not a block having a predetermined size constituting a video satisfies a condition corresponding to the skip mode among a plurality of encoding modes used for video encoding. The skip mode condition is as described above. If the skip mode condition is satisfied, the encoding mode determination unit 920 determines the skip mode as the encoding mode of the block, and if not satisfied, further performs interpretation or intra-prediction. Then, one of the various encoding modes is determined as the encoding mode of the block.

  The encoding execution unit 930 generates compressed moving image data according to the encoding mode determined by the encoding mode determination unit 920.

  FIG. 10 shows another embodiment of the encoding apparatus of the present invention.

  The temporal similarity calculation unit 1010 calculates the temporal similarity between the macroblock in the current picture and the reference block of the reference picture under the inter mode.

  The comparison unit 1020 determines whether to omit the encoding process under the intra mode based on the difference between the measured temporal similarity and the spatial similarity between the macroblock and the adjacent block in the current picture. In this way, one encoding mode is determined. The temporal similarity and the spatial similarity are preferably calculated based on the average rate and the average boundary error ABE described above.

  Yet another example of determining temporal similarity is to use the pixel value difference between the macroblock in the current picture and the reference block of the reference picture under inter mode. This difference is calculated by calculating the difference between each pixel value constituting the current macroblock and the pixel value in the reference block of the reference picture corresponding to the position of each pixel, and then obtaining the sum or square sum of those absolute values. And then multiplied by a predetermined ratio λ.

  The encoding execution unit 1030 generates compressed moving image data according to the encoding mode determined by the comparison unit 1020.

  On the other hand, the encoding method described above can be created by a computer program. The code and code segments that make up the program can be easily inferred by computer programmers in the art. The program is stored in a computer-readable information recording medium, and is read and executed by the computer to implement the encoding method. The information recording medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

  The present invention has been mainly described with reference to preferred embodiments. Those skilled in the art will understand that the present invention may be embodied in a modified form without departing from the essential characteristics of the present invention. Accordingly, the disclosed embodiments should be considered in an illustrative, not a limiting sense. The scope of the present invention is shown not in the above description but in the claims, and all differences within the equivalent scope should be construed as being included in the present invention.

  According to the present invention, video encoding can be performed quickly without a significant decrease in image quality. That is, the loss of SNR is not large, and the number of times of calculation of RDcost can be reduced by 70% or more, and the time required for video encoding can be shortened by about 30%.

1 is a block diagram of an encoding apparatus according to a preferred embodiment of the present invention. It is a figure which shows the variable block which a macroblock can have at the time of interpretation. It is a figure which shows an example of the block shape allocated to each macroblock of the present picture which encodes. It is a figure which shows the luminance block P which it is going to predict, and the adjacent block used for prediction. It is a figure for demonstrating calculation of PMV. It is a figure for demonstrating calculation of PMV. It is a figure for demonstrating calculation of RDcost. It is a flowchart showing the method of determining encoding mode. It is a figure for demonstrating calculation of the average boundary error ABE. It is a figure for demonstrating calculation of the average boundary error ABE. It is the block diagram which simplified the encoding apparatus of this invention. It is another Example of the encoding apparatus of this invention.

Claims (23)

In the video encoding method,
(A) It is determined whether a block having a predetermined size constituting the video among a plurality of encoding modes used for encoding the video satisfies a condition corresponding to a predetermined first mode. Steps,
And (b) a step of determining the first mode as the encoding mode of the block when the predetermined condition is satisfied.   The encoding method according to claim 1, further comprising: (c) generating encoded moving image data for the block under the first mode. The encoding method according to claim 1, wherein the first mode is a skip mode in which only the encoding mode information of the block is transmitted or stored.   The first mode is H.264. The encoding method according to claim 3, wherein the skip mode is H.264.   The predetermined condition is that a block size used for motion estimation of the predetermined block is 16 × 16, and a reference picture is a picture to which the predetermined block belongs among reference pictures stored in a frame memory. And the motion vector is (0, 0) or the same as the predicted motion vector, and the difference between pixel values of the predetermined block and the reference picture block is converted. 2. The encoding method according to claim 1, wherein the quantized results are all zero.   The predetermined condition is that a block size used for estimation of motion of the predetermined block is 8 × 8, and the first reference picture and the first motion vector are the first and second spatial blocks of the block. 2 is the same as that estimated from the reference picture and the second motion vector information, and the result of transforming and quantizing the difference between the pixel values of the predetermined block and the adjacent block of the reference picture is 0. The encoding method according to claim 1, characterized in that: In the video encoding method,
(A) calculating a temporal similarity with a reference block of a reference picture under an inter mode for a block of a predetermined size in a current picture constituting the video;
(B) The step of omitting the encoding process under the intra mode according to the calculated temporal similarity and the difference in spatial similarity between the block and the block adjacent to the current picture is included. An encoding method characterized by the above.   The temporal similarity is determined by the number of bits required to encode the block, and the spatial similarity is determined by a difference in pixel values between a current picture block to be encoded and an adjacent block. The encoding method according to claim 7, wherein:   The temporal similarity is determined by a difference in pixel value between the block and a reference block of a reference picture, and the spatial similarity is determined by a pixel value between a current picture block to be encoded and an adjacent block. The encoding method according to claim 7, wherein the encoding method is determined by the difference. The spatial similarity is
Within the block to be encoded, the difference in pixel value between the pixel existing at the boundary with another block and the pixel of the decoded block adjacent to the pixel is calculated for each luminance and chromaticity. The encoding method according to claim 8, wherein the encoding method is determined as follows. The temporal similarity is
The number of bits required for encoding the difference value between the current picture block and the reference block corresponding thereto, the number of bits required for encoding the motion vector, and the number of bits required for encoding the reference picture information The encoding method according to claim 8, wherein the encoding method is determined by information to be included. A determination unit configured to determine whether or not a block having a predetermined size that constitutes the video among a plurality of encoding modes used for video encoding satisfies a condition corresponding to a predetermined first mode;
An encoding apparatus comprising: an encoding mode determining unit that determines the first mode as the encoding mode of the block when the predetermined condition is satisfied.   The encoding apparatus according to claim 12, further comprising an encoding execution unit that generates moving image data compressed under the first mode. The first mode is:
The encoding apparatus according to claim 12, wherein the encoding mode is a skip mode in which only the encoding mode information of the block is transmitted or stored.   The predetermined condition is that a block size used for motion estimation of the predetermined block is 16 × 16, and a reference picture is a picture to which the predetermined block belongs among reference pictures stored in a frame memory. And the motion vector is (0, 0) or the same as the predicted motion vector, and the difference between pixel values of the predetermined block and the reference picture block is converted. 13. The encoding apparatus according to claim 12, wherein the quantized results are all zero.   The predetermined condition is that a block size used for estimation of motion of the predetermined block is 8 × 8, and the first reference picture and the first motion vector are the first and second spatial blocks of the block. 2 is the same as that estimated from the reference picture and the second motion vector information, and the result of transforming and quantizing the difference between the pixel values of the predetermined block and the adjacent block of the reference picture is 0. The encoding device according to claim 12, characterized in that: A temporal similarity calculation unit that calculates temporal similarity with a reference block of a reference picture under an inter mode for a block of a predetermined size in a current picture;
A comparison in which one encoding mode is determined by omitting the encoding process under the intra mode according to the measured temporal similarity and the difference in spatial similarity between the block and the block adjacent to the current picture. An encoding device.   The encoding apparatus according to claim 17, further comprising an encoding execution unit that generates moving image data compressed under the predetermined encoding mode.   The temporal similarity is determined by the number of bits required for encoding the block, and the spatial similarity is determined by a difference in pixel values between the predetermined block to be encoded and a block adjacent thereto. 18. The encoding apparatus according to claim 17, wherein the encoding apparatus is determined.   The temporal similarity is determined by a pixel value difference between the block and a reference block of a reference picture, and the spatial similarity is a pixel value between the predetermined block to be encoded and a block adjacent thereto. The encoding apparatus according to claim 17, wherein the encoding apparatus is determined by a difference between The spatial similarity is
Within a predetermined block to be encoded, the difference in pixel value between a pixel existing at a boundary with another block and a pixel of a decoded block adjacent to the pixel is classified by luminance and chromaticity, respectively. The encoding apparatus according to claim 19, wherein the encoding apparatus is determined by calculation. The temporal similarity is
The number of bits required for encoding the difference value between the current picture block and the reference block corresponding thereto, the number of bits required for encoding the motion vector, and the number of bits required for encoding the reference picture information The encoding apparatus according to claim 19, wherein the encoding apparatus is determined by information to be included. In a computer-readable medium in which a program for performing a video encoding method is implemented, the video encoding method includes:
(A) It is determined whether a block having a predetermined size constituting the video among a plurality of encoding modes used for encoding the video satisfies a condition corresponding to a predetermined first mode. Steps,
(B) A computer-readable medium comprising: a step of determining the first mode as an encoding mode of the block when the predetermined condition is satisfied.

Images