JP6985915B2 - Decoding error detection and estimation device, video decoding device, and their programs - Google Patents

Description

The present invention relates to a technique for detecting a decoding error during decoding of a coded video, and in particular, a decoding error detection and estimation device for detecting or estimating a decoding error generated when decoding a coded video stream, and detecting the decoding error. The present invention relates to a video decoding device that performs error concealment according to the above, and these programs.

Generally, in the video coding method, a specific bit string called a start code is specified. The start code is used to separate the bit stream. The start code is guaranteed to appear only at the delimiter position.

The part other than the start code separated by the start code is referred to as a payload in the present specification (the payload is referred to as a NAL unit in the video coding method according to H.264 or H.265 or the like). The payload is composed of a header part showing information such as slices and macroblocks, and a data part showing orthogonal transformation coefficients, motion vectors, and their residual signals. Except for streams with extremely low bit rates, almost all bits in the payload generally belong to the data part. Further, the area of the image obtained by decoding the payload is referred to as a decoding area (that is, an area to be allocated to the frame image for restoring the image obtained by decoding the payload) in the present specification. The decoded area is the entire screen (called a frame image, a picture, etc.) or a part of the screen (called a slice, a tile, etc.), and is specified by the stream header.

In recent video coding methods such as H.265 / MPEG-H HEVC method, in order to improve bit usage efficiency, an unspecified bit string that makes it difficult to detect decoding errors appears in the payload other than the start code (especially in data). do. Therefore, the known method of detecting the decoding error that occurred when decoding the encoded video stream is that the start code appears before the decoding of the decoding area is completed (payload is insufficient), or the decoding of the decoding area is completed. Despite this, it is based on the detection that the start code does not appear (payload surplus). In the known decoding error detection technique using this start code, since the error position in the decoding area cannot be grasped, the entire decoding area must be the target of error concealment, and therefore the processing efficiency is not good.

A decoding error detection circuit that detects an error when undecidable data is generated during decoding of variable length data is known (see, for example, Patent Document 1). This variable-length data is data compressed by variable-length coding in which data represented by Huffman coding and codewords have a one-to-one correspondence. Since only a specific bit combination exists in the codeword used in variable-length coding, decoding error detection is possible.

However, recent video coding methods such as the H.265 / MPEG-H HEVC method use arithmetic coding instead of variable length coding (especially in the data section) in order to improve bit usage efficiency. Since there is no predetermined code word in arithmetic coding, an unspecified bit string that makes it difficult to detect a decoding error appears in a payload other than the start code (particularly in data) as described above. Therefore, Patent Document 1 Decoding error detection cannot be performed with this technique.

Further, a technique for detecting a macroblock in which a decoding error has occurred during video decoding is also known (see, for example, Patent Document 2). Patent Document 2 discloses the following two examples as a technique for detecting a decoding error.

(1) As a result of decoding the macroblock information, the corresponding data does not exist. This is, for example, when the macroblock information is decoded and then reproduced by referring to a certain reference frame, the reference frame points to "List 10", and the reference frame list changes from "List 0" to "List". For example, when only up to 8 ”existed.

(2) The value obtained by decoding the macroblock information does not exist in the coding correspondence table. This is the case, for example, when the decoded result is a value of "40", but only a value from "0" to "30" is defined in the coding correspondence table.

However, since the technique disclosed in Patent Document 2 is a technique for detecting a decoding error of a macroblock related to a header part in which a reference frame list or a coded correspondence table is shown, most of the bit strings in the payload are generally the data part. Since it is (especially the orthogonal conversion coefficient or its residual signal), a high detection probability cannot be expected.

In recent video coding methods such as H.265 / MPEG-H HEVC method, in order to maximize the bit usage efficiency, values that do not exist in the coding correspondence table are not decoded in the first place. It is difficult to detect a decoding error by the technique disclosed in Document 2.

Japanese Unexamined Patent Publication No. 5-95536 Japanese Patent No. 43097884

As mentioned above, recent video coding methods such as the H.265 / MPEG-H HEVC method make it difficult to detect decoding errors in the payload other than the start code (especially in the data) in order to improve bit usage efficiency. A specific bit string appears. Therefore, the known method of detecting the decoding error that occurred when decoding the encoded video stream is that the start code appears before the decoding of the decoding area is completed (payload is insufficient), or the decoding of the decoding area is completed. Despite this, it is based on the detection that the start code does not appear (payload surplus). In the known decoding error detection technique using this start code, since the error position in the decoding area cannot be grasped, the entire decoding area must be the target of error concealment, and therefore the processing efficiency is not good.

Further, in the technique of Patent Document 1 or 2, it is difficult to detect a decoding error generated when decoding a coded video stream according to a recent video coding method such as H.265 / MPEG-H HEVC method. ..

Therefore, H. Not only in the video coding method before 264, but also in the recent video coding method such as H.265 / MPEG-H HEVC method, the decoding error generated when decoding the coded video stream can be efficiently detected or detected. A technique for estimating is desired.

Therefore, an object of the present invention is H. Not only in the video coding method before 264, but also in the recent video coding method such as H.265 / MPEG-H HEVC method, the decoding error generated when decoding the coded video stream can be efficiently detected or detected. It is an object of the present invention to provide a decoding error detection estimation device for estimating, a video decoding device for performing error concealment according to the detection of the decoding error, and a program thereof.

The decoding error detection and estimation device of the first aspect according to the present invention is a decoding error detection and estimation device that detects or estimates a decoding error that occurs when decoding a coded video stream, and decodes the coded video stream. A clip number detecting means for inputting the obtained decode area and detecting the number of clips in the decode error determination block of the block size preset for performing the decode error determination for the input decode area, and the detected decode error. A predetermined judgment criterion is obtained by comparing the number of clips of the current block constituting the judgment block with the number of clips of the past block (block in the past frame) at the same block position as the current block in the predetermined past frame to be compared. According to the above, the present block is provided with a decoding error determining means for determining whether or not the current block has a decoding error and generating a decoding error determining signal indicating the presence or absence of the decoding error.

Further, in the decode error detection and estimation device of the first aspect according to the present invention, the decode error determination means has a predetermined amount of increase in the number of clips in the current block from the number of clips in the past block as the predetermined determination criterion. It is characterized in that it is determined that there is a decoding error when it is determined.

Further, in the decoding error detection and estimation device of the first aspect according to the present invention, the decoding error determining means sets the past frame that has been decoded immediately before in the decoding order as the comparison target as the predetermined past frame of the comparison target. It is characterized by.

Further, the decoding error detection and estimation device of the second aspect according to the present invention is a decoding error detection and estimation device that detects or estimates a decoding error that occurs when decoding a coded video stream, and decodes the coded video stream. The number of clips detecting means for detecting the number of clips in the decoding error determination block of the block size preset for inputting the decoded area obtained by the above and performing the decoding error determination for the input decoding area, and the detection. Decoding error determination The number of clips in the current block that constitutes the block is compared with a predetermined value, and it is determined whether or not the current block has a decoding error according to a predetermined determination criterion, and the presence or absence of a decoding error is indicated. It is characterized by comprising a decoding error determining means for generating a decoding error determining signal.

Further, in the decode error detection and estimation device of the second aspect according to the present invention, the decode error determination means compares the specified value with the number of clips in the current block as the predetermined determination criterion, and the current block. When it is determined that the number of clips in the above is increased by a predetermined amount from the specified value, it is determined that there is a decoding error.

Further, in the decoding error detection and estimation apparatus of the first and second aspects according to the present invention, the decoding error determining means is characterized in that the decoding error determination means is compared and determined by the ratio of the number of clips to the number of pixels in the decoding error determination block. do.

Further, the video decoding device of the present invention is an error concealment by complementing the corresponding decoding area based on the decoding error detection and estimation device according to the present invention and the decoding error determination signal obtained from the decoding error detection and estimation device. It is characterized by providing a complementary means for generating a mented video stream.

Further, the program of the present invention is configured as a program for causing the computer to function as the decoding error detection and estimation device according to the present invention or the video decoding device according to the present invention.

According to the present invention, in order to detect or estimate a decoding error generated when decoding a coded video stream, the number of clips of pixel values generated in the decoding process is used. Efficiently detect or estimate the location where a decoding error has occurred not only in the video coding method before 264 but also in the recent video coding method with high bit usage efficiency such as H.265 / MPEG-H HEVC method. Is possible.

It is a block diagram which illustrates the schematic structure of the video decoding apparatus which comprises the decoding error detection estimation apparatus of 1st Embodiment by this invention. It is explanatory drawing which shows an example of the block division which concerns on the decoding error determination block in the decoding error detection estimation apparatus of 1st Embodiment by this invention. It is explanatory drawing which shows the comparative example of the number of clips used for the decoding error determination in the decoding error detection estimation apparatus of 1st Embodiment by this invention. It is a block diagram which illustrates the schematic structure of the video decoding apparatus which comprises the decoding error detection estimation apparatus of 3rd Embodiment by this invention. It is explanatory drawing of the number of clips used for the decoding error determination in the decoding error detection estimation apparatus of 2nd Embodiment by this invention.

[First Embodiment]
(Video decoder)
FIG. 1 is a block diagram illustrating a schematic configuration of a video decoding device 1 including the decoding error detection and estimation device 10 of the first embodiment according to the present invention. The decoding error detection and estimation device 10 according to the present invention functions as a device for detecting or estimating a decoding error generated when decoding a coded video stream.

First, the video decoding device 1 includes a decoding error detection and estimation device 10, a decoding processing unit 11, and a complementary unit 12.

The decoding processing unit 11 is a functional unit that inputs a coded video stream and performs decoding processing. When the coded video stream is input, the decoding processing unit 11 can classify the decoding area specified by the stream header at the time of decoding, and sequentially outputs the decoding area to the complement unit 12.

In general, H. In the video decoding device in recent video coding methods such as H.265 / MPEG-H HEVC method as well as the video coding method before 264, it is within the allowable range at the time of decoding (when the decoding area is generated). And the process of clipping the pixel value is included. That is, the clip is a process of replacing the pixel value exceeding the allowable minimum value or maximum value with the allowable minimum value or maximum value, respectively. Generally, in the video coding method, it is specified that clipping is performed in the decoding process (for example, in the H.265 / MPEG-H HEVC method, it is defined as a function such as Clip3). Also, the permissible range (minimum value or maximum value) is not fixed but may be variable in the video stream. The clip in this case is a process of clipping the pixel value within the range allowed by the block.

The decoding error detection and estimation device 10 of the first embodiment first inputs a decoding area designated as a determination target of a decoding error in real time or from the decoding processing unit 11, and determines the decoding error for the input decoding area. The block is divided into decode error judgment blocks with a preset block size. Subsequently, the decode error detection and estimation device 10 detects the number of clips in the current block constituting the decode error determination block, and the number of clips in the detected current block and the predetermined past frame to be compared (preferably, the past frame). Compare the number of clips in the past block at the same block position (same block size) with the current block in the past frame that has been decoded immediately before in the decoding order. Then, the decode error detection and estimation device 10 determines whether or not the current block has a decode error according to a predetermined determination criterion when comparing the number of clips of the current block with the number of clips of the past block. , A decoding error determination signal indicating the presence or absence of a decoding error (including indicating only that there is a decoding error or indicating only that there is no decoding error) is generated and output to the complement unit 12 or to the outside.

The complement unit 12 sequentially temporarily stores the decode area input from the decode processing unit 11, and complements the corresponding decode area based on the decode error determination signal obtained from the decode error detection and estimation device 10 to obtain an error. Generates a concealed video stream and outputs it to the outside. It should be noted that the technique itself relating to this complement is not the gist of the present invention, but may be a known technique by interpolation, extrapolation, or the like.

Although the decoding error detection and estimation device 10 of the present embodiment shown in FIG. 1 shows an example of incorporating the decoding error detection and estimation device 10 into the video decoding device 1, the decoding error detection and estimation device 10 of the present embodiment is placed outside the video decoding device 1. It may be provided. Further, the decode error detection and estimation device 10 of the present embodiment can be configured simply as a device that outputs a decode error determination signal for the decode region to the outside. For example, it can be used for imparting a decode error determination signal (including indicating only with a decode error or indicating only without a decode error) to the decode area in the stored moving image file.

(Decoding error detection and estimation device)
Hereinafter, the decoding error detection and estimation device 10 of the present embodiment shown in FIG. 1 will be described in detail. As shown in FIG. 1, the decode error detection and estimation device 10 of the present embodiment includes a target decode area extraction unit 101, a block division unit 102, a clip number detection unit 103, a clip memory 104, and a decode error determination unit 105. Further, the decoding error determination unit 105 includes a clip number comparison unit 105a.

The target decoding area extraction unit 101 inputs the decoding area in real time from the decoding processing unit 11, or extracts and inputs the decoding area designated as the determination target of the decoding error, and inputs the input decoding area into the block division unit 102. Output to.

The block division unit 102 divides a block into a decoding error determination block having a block size set in advance for performing a decoding error determination on the input decoding area obtained from the target decoding area extraction unit 101, and each decoding error determination block. Is output to the clip number detection unit 103. Further, the block division unit 102 outputs the frame number of the frame image including the decode area, the block size of each decode error determination block, and the block division information indicating the block position to the clip number detection unit 103 and the decode error determination unit 105. ..

The block size of each decoding error determination block that divides the decoding area can be any size. FIG. 2 is an explanatory diagram showing an example of block division relating to a decoding error determination block in the decoding error detection and estimation device 10 of the first embodiment according to the present invention. For example, as shown in FIG. 2, it is assumed that the decoding area DI is composed of a part (slice, tile, etc.) of the nth frame image F (n) in the decoding order in the encoded video stream (note that the decoding area DI is composed of a part thereof (slices, tiles, etc.). The decode area DI may be the entire frame and is specified in the stream header). The decoding area DI is divided by the block size of the basic unit for encoding / decoding, assuming that the kth coding / decoding block is bk # k, but the mth decoding error determination block is BK # m. Then, the block size can be made larger than the block size of the coding / decoding block bk # k.

The larger the block size of the decode error determination block BK # m, the more false detection and omission of detection can be prevented and the processing speed is improved. become. From this point of view, as the block size of the decoding error determination block BK # m, the coding / decoding block, which is the basic unit for encoding / decoding, is the block size of bk # k (for example, in H.265 / MPEG-H HEVC). It is preferable that the size is the same as that of CTU).

The clip number detection unit 103 cumulatively detects the number of clips in each decode error determination block based on the block division information obtained from the block division unit 102 and each decode error determination block, and detects the number of clips in each decode error determination block. Is output to the decoding error determination unit 105 as the number of clips in the current block. Further, the clip number detection unit 103 outputs the number of clips of the current block detected based on the block division information to the clip memory 104. Instead of configuring the block division unit 102 to output the block division information to the decode error determination unit 105, the clip number detection unit 103 outputs the block division information received from the block division unit 102 to the decode error determination unit 105. It can also be configured to be used.

The clip memory 104 temporarily determines the number of clips of the current block obtained from the clip number detection unit 103 based on the block division information, the frame number of the frame image including the decode area, the block size of each decode error determination block, and the block position. accumulate. The clip memory 104 updates and holds the number of clips of each decoding error determination block corresponding to the predetermined number of frames. In this example, each decoding error determination block in the past frame that has been decoded immediately before in the decoding order is used. Update and keep the number of clips. The clip memory 104 holds no clips (zero number of clips) in the initial state. Instead of the frame number, the clip memory 104 may hold information that can identify the current frame, for example, a relative value from the beginning of GOP (Group Of Pictures) or an index in a predetermined coding order.

The decoding error determination unit 105 is a past frame (this example) designated in advance based on the block division information obtained from the block division unit 102 or the clip number detection unit 103 and the number of clips in the current block obtained from the clip number detection unit 103. Then, the number of clips of the past block at the same block position (same block size) as the current block in the past frame that has been decoded immediately before in the decoding order is acquired from the clip memory 104.

Then, the decode error determination unit 105 compares the number of clips in the current block with the number of clips in the past block according to a predetermined determination criterion by the clip number comparison unit 105a, and the current block has a decode error. It determines whether or not it is present, and generates a decode error determination signal (including indicating only with a decode error and indicating only without a decode error) indicating the presence or absence of a decode error according to the result of the clip number comparison unit 105a. Then, it is output to the complementary unit 12 or to the outside.

FIG. 3 is an explanatory diagram showing a comparative example of the number of clips used for decoding error determination in the decoding error detection and estimation device 10 of the first embodiment according to the present invention. As shown in FIG. 3, the clip number comparison unit 105a specifies in advance the number of clips in the current block BK # m of the decoding error determination target in the nth frame image F (n) in the decoding order in the encoded video stream. The number of clips of the past block BK # m'at the same block position (same block size) as the current block in the past frame (in this example, the past frame F (n-1) decoded immediately before in the decoding order). The presence or absence of a decoding error is determined by comparing according to a predetermined criterion.

Here, the clip number comparison unit 105a compares the number of clips in the past block with the number of clips in the current block as an example of the predetermined determination criterion, and the number of clips in the current block is higher than the number of clips in the past block. When it is determined that there is an increase in the fixed amount, it is determined that there is a decoding error. Preferably, the determination is made by comparing the number of clips with respect to the number of pixels in the decoding error determination block.

For example, as an example of the predetermined determination criterion, the increase in the predetermined amount of clips means that the number of clips in the past block is P, the number of clips in the current block is C, and the number of pixels in the decoding error determination block is S. , Can be configured as in the following three examples.
(1) When (C + 1) / (P + 1)> k is satisfied, it is determined that there is a decoding error.
(2) When C / S-P / S> n is satisfied, it is determined that there is a decoding error.
(3) When C / S-P * k / S> n is satisfied, it is determined that there is a decoding error.

It should be noted that k and n in the above equation are parameters to be set in advance. Set a value of 1.0 or more for k and a value of 0 or more for n. The smaller the value of k or n, the smaller the omission of detection of the decoding error, but the larger the false detection. On the contrary, the larger the value of k or n, the smaller the false detection of the decoding error, but the larger the detection omission. For example, when the criterion of (2) above is used, it is preferable to set the value of n to about 0.01 to 0.03.

According to the decoding error detection and estimation device 10 of the first embodiment and the video decoding device 1 including the same, the decoding error is generated in the decoding process in order to detect or estimate the decoding error generated when the encoded video stream is decoded. Since the number of clips of the pixel value to be used is used, H. In recent video coding methods with high bit usage efficiency such as H.265 / MPEG-H HEVC method as well as video coding methods before 264, it is possible to efficiently detect or estimate the location where a decoding error has occurred. It will be possible.

[Second Embodiment]
(Video decoder)
FIG. 4 is a block diagram illustrating a schematic configuration of a video decoding device 1 including the decoding error detection and estimation device 10 of the second embodiment according to the present invention. The same reference numbers are assigned to the same components as those in the first embodiment shown in FIG. The video decoding device 1 according to the second embodiment is the same except that the configuration of the decoding error detection and estimation device 10 is different from that of the first embodiment. The explanation of is omitted. Further, the decoding error detection and estimation device 10 of the second embodiment is different from the first embodiment in that the decoding error is determined without holding the number of clips of the past block in the clip memory 104.

Although the decoding error detection and estimation device 10 of the present embodiment shown in FIG. 4 shows an example of incorporating the decoding error detection and estimation device 1 into the video decoding device 1, the decoding error detection and estimation device 10 of the present embodiment is the same as the first embodiment. May be provided outside the video decoding device 1. Further, the decode error detection and estimation device 10 of the present embodiment can be configured simply as a device that outputs a decode error determination signal for the decode region to the outside. For example, it can be used for imparting a decode error determination signal (including indicating only with a decode error or indicating only without a decode error) to the decode area in the stored moving image file.

(Decoding error detection and estimation device)
Hereinafter, the decoding error detection and estimation device 10 of the present embodiment shown in FIG. 4 will be described in detail. As shown in FIG. 4, the decode error detection and estimation device 10 of the present embodiment includes a target decode area extraction unit 101, a block division unit 102, a clip number detection unit 103, and a decode error determination unit 105. Further, the decoding error determination unit 105 includes a clip number comparison unit 105b.

Similar to the first embodiment, the target decoding area extraction unit 101 inputs the decoding area in real time from the decoding processing unit 11, or extracts and inputs the decoding area designated as the determination target of the decoding error, and inputs the decoding area. The decoded area is output to the block division unit 102.

Similar to the first embodiment, the block division unit 102 blocks to a decoding error determination block having a block size set in advance for performing decoding error determination on the input decoding area obtained from the target decoding area extraction unit 101. It is divided and each decoding error determination block is output to the clip number detection unit 103. Further, the block division unit 102 outputs the frame number of the frame image including the decode area, the block size of each decode error determination block, and the block division information indicating the block position to the clip number detection unit 103 and the decode error determination unit 105. ..

Also in this embodiment, as described with reference to FIG. 2 for the first embodiment described above, the block size of each decoding error determination block that divides the decoding area can be any size.

Similar to the first embodiment, the clip number detection unit 103 cumulatively detects the number of clips in each decode error determination block based on the block division information obtained from the block division unit 102 and each decode error determination block. The number of clips in each decode error determination block is output to the decode error determination unit 105 as the number of clips in the current block. Instead of configuring the block division unit 102 to output the block division information to the decode error determination unit 105, the clip number detection unit 103 outputs the block division information received from the block division unit 102 to the decode error determination unit 105. It can also be configured to be used.

The decode error determination unit 105 is determined by the clip number comparison unit 105b based on the block division information obtained from the block division unit 102 or the clip number detection unit 103 and the number of clips in the current block obtained from the clip number detection unit 103. The number of clips of the current block is compared with a predetermined predetermined value according to the determination criteria of the above, it is determined whether or not the current block has a decoding error, and according to the result of the clip number comparison unit 105b, A decoding error determination signal indicating the presence or absence of a decoding error (including indicating only with a decoding error or indicating only without a decoding error) is generated and output to the complement 12 or to the outside.

FIG. 5 is an explanatory diagram showing a comparative example of the number of clips used for decoding error determination in the decoding error detection and estimation device 10 of the second embodiment according to the present invention. As shown in FIG. 5, the clip number comparison unit 105b determines in advance the number of clips in the current block BK # m of the decoding error determination target in the nth frame image F (n) in the decoding order in the encoded video stream. The presence or absence of a decoding error is determined by comparing the specified values according to a predetermined criterion.

Here, the clip number comparison unit 105b compares a specified value (for example, a predetermined number of clips) with the number of clips in the current block as an example of the predetermined determination standard, and the number of clips in the current block is the specified value. When it is determined that there is a predetermined amount of increase from the value, it is determined that there is a decoding error. Preferably, the determination is made by comparing the number of clips with respect to the number of pixels in the decoding error determination block.

For example, as an example of the predetermined determination standard, an increase in the predetermined amount of clips means that the number of clips in the current block is C and the number of pixels in the decoding error determination block is S, as in the following two examples. Can be configured.
(4) When C> n is satisfied, it is determined that there is a decoding error.
(5) When C / S> n is satisfied, it is determined that there is a decoding error.

Note that n in the above equation is a parameter to be set in advance. The smaller the value of n, the smaller the omission of detection of decoding errors, but the more false detections. On the contrary, as the value of n is larger, the false detection of the decoding error is reduced, but the detection omission increases. For example, when the criterion of (5) above is used, it is preferable to set the value of n to about 0.01 to 0.03.

Also in the second embodiment, in order to detect or estimate the decoding error generated when decoding the coded video stream, the number of clips of the pixel value generated in the decoding process is used. In recent video coding methods with high bit usage efficiency such as H.265 / MPEG-H HEVC method as well as video coding methods before 264, it is possible to efficiently detect or estimate the location where a decoding error has occurred. It will be possible.

Further, in the second embodiment, only the number of clips in the current block needs to be detected, so that the processing load can be reduced as compared with the first embodiment.

The decoding error detection and estimation device 10 or the video decoding device 1 in each of the above embodiments can be configured by a computer, and is a program for operating each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1. Can be preferably used. Specifically, the control unit for controlling each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1 can be configured by the central processing unit (CPU) in the computer, and each processing unit is operated. At least one memory can be used as a storage unit for appropriately storing a program required for making the program. That is, by causing such a computer to execute the program by the CPU, it is possible to realize the functions of each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1. Further, a program for realizing the functions of each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1 can be stored in a predetermined area of the above-mentioned storage unit (memory). Such a storage unit can be configured by a RAM or ROM inside the device, or can be configured by an external storage device (for example, a hard disk). Further, such a program can be configured as a part of software (stored in ROM or an external storage device) on an OS used by a computer. Further, a program for causing such a computer to function as each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1 can be recorded on a computer-readable recording medium. Further, each processing unit of the decoding error detection and estimation device 10 or the video decoding device 1 can be configured as a part of hardware or software, and each of them can be combined and realized.

Although the present invention has been described above with reference to examples of specific embodiments, the present invention is not limited to the examples of the above-described embodiments, and can be variously modified without departing from the technical idea. For example, in the example of the above-described embodiment, specific determination criteria have been exemplified and described, but various modifications such as the combined use of squared averaging are assumed. Therefore, the transmitting device and the receiving device according to the present invention are not limited to the examples of the above-described embodiments, but are limited only by the description of the scope of claims.

According to the present invention, a decoding error generated when decoding a coded video stream can be efficiently detected or estimated, which is useful for applications of a video coding method requiring error concealment.

1 Video decoding device 10 Decoding error detection and estimation device 11 Decoding processing unit 12 Complementary unit 101 Target decoding area extraction unit 102 Block division unit 103 Clip number detection unit 104 Clip memory 105 Decoding error determination unit 105a, 105b Clip number comparison unit

Claims (8)

A decoding error detection and estimation device that detects or estimates decoding errors that occur when decoding a coded video stream.
The number of clips for inputting the decoding area obtained by decoding the encoded video stream and detecting the number of clips in the decoding error determination block of the block size preset for performing the decoding error determination for the input decoding area. Detection means and
The number of clips of the current block constituting the detected decode error determination block is compared with the number of clips of the past block at the same block position as the current block in the predetermined past frame to be compared, and the relevant determination criteria are followed. A decoding error determination means that determines whether or not the current block has a decoding error and generates a decoding error determination signal indicating the presence or absence of a decoding error.
A decoding error detection and estimation device. The decoding error determining means is characterized in that, as the predetermined determination criterion, when it is determined that the number of clips in the current block is increased by a predetermined amount from the number of clips in the past block, it is determined that there is a decoding error. The decoding error detection and estimation device according to claim 1. The decoding error detection according to claim 1 or 2, wherein the decoding error determining means targets a past frame that has been decoded immediately before in the decoding order as a predetermined past frame for comparison. Estimator. A decoding error detection and estimation device that detects or estimates decoding errors that occur when decoding a coded video stream.
The number of clips for inputting the decoding area obtained by decoding the encoded video stream and detecting the number of clips in the decoding error determination block of the block size preset for performing the decoding error determination for the input decoding area. Detection means and
The number of clips of the current block constituting the detected decode error determination block is compared with a predetermined predetermined value, and it is determined whether or not the current block has a decode error according to a predetermined determination criterion. Decoding error judgment means that generates a decoding error judgment signal indicating the presence or absence, and
A decoding error detection and estimation device. The decoding error determining means compares the specified value with the number of clips in the current block as the predetermined determination criterion, and determines that the number of clips in the current block is increased by a predetermined amount from the specified value. The decoding error detection and estimation device according to claim 4, wherein it is determined that there is a decoding error. The decode error detection estimation according to any one of claims 1 to 5, wherein the decode error determination means makes a determination by comparing the number of clips with respect to the number of pixels in the decode error determination block. Device. The decoding error detection and estimation device according to any one of claims 1 to 6.
Complementing means for generating an error-concealed video stream by complementing the corresponding decoding region based on the decoding error determination signal obtained from the decoding error detection and estimation device.
A video decoding device characterized by comprising. A program for making a computer function as the decoding error detection and estimation device according to any one of claims 1 to 6 or the video decoding device according to claim 7.

Images