CN-122029816-A - Video encoding/decoding method and device based on intra-picture prediction

Abstract

The present invention relates to an improved encoding/decoding method and apparatus based on Intra Block Copy (IBC) in an intra prediction encoding/decoding method. A decoding method of an encoded video bitstream according to an embodiment of the present invention may include a step of reading prediction mode information from the bitstream, wherein the prediction mode information is used for predictive decoding (prediction decoding) of a current block (current block) of a picture being currently decoded, a step of determining whether the prediction mode information represents a mode based on Intra Block Copy (IBC), a step of reading at least one prediction vector (prediction vector) from the bitstream based on the prediction mode information, wherein the prediction vector is used for specifying a position of at least one reference block (REFERENCE BLOCK) from at least one picture including the picture being currently decoded, a step of obtaining at least one reference block (REFERENCE BLOCK) based on the prediction vector, a step of generating a prediction block (prediction block) based on the at least one reference block, and a step of performing predictive decoding of the current block based on the prediction block.

Inventors

• LIU CHANGYOU
• LI SHANYING

Assignees

• 佳稳集团有限公司
• 株式会社迓廷试

Dates

Publication Date

20260512

Application Date

20241017

Priority Date

20231019

Claims (20)

1. 1. A method of decoding an encoded video bitstream, comprising: A step of reading prediction mode information from the bitstream, wherein the prediction mode information is used for predictive decoding of a current block of a current image being decoded; A step of judging whether the prediction mode information indicates a mode of copying IBCs based on intra blocks; a step of reading at least one prediction vector from the bitstream based on the prediction mode information, wherein the prediction vector is used to specify a position of at least one reference block from at least one picture including the picture currently being decoded; a step of obtaining at least one reference block based on the prediction vector; a step of generating a prediction block based on the at least one reference block, and And performing predictive decoding of the current block based on the predicted block.
2. 2. The decoding method of claim 1, wherein, A step of reading the prediction vector, comprising: a step of obtaining an adaptive block vector resolution ABVR flag value from the bitstream; Reading precision expression information of the block vector when the ABVR flag value corresponds to a first range; A step of determining the precision of the prediction vector based on the ABVR flag information and the precision expression information, and And reading the prediction vector based on the resolution.
3. 3. The decoding method of claim 2, wherein, The precision of the prediction vector is determined by one of at least one precision unit comprising sub-pixel units.
4. 4. The decoding method of claim 3, wherein, Determining the precision of the prediction vector as a 1/4 pixel unit when the ABVR flag value does not correspond to a first range; Determining the precision of the prediction vector as a 1-pixel unit when the ABVR flag value corresponds to a first range and the value of the precision expression information is a first value; when the ABVR flag value corresponds to a first range and the value of the precision expression information is a second value, the precision of the prediction vector is determined to be a 4-pixel unit.
5. 5. The decoding method according to any one of claims 2 to 4, wherein, The first range refers to the ABVR flag value being "1", The precision expression information is an index value having at least two cases.
6. 6. The decoding method of claim 1, further comprising: A step of judging whether or not the prediction mode information indicates a redefined mode using a prediction vector, and A step of redefining the prediction vector by performing compensation correction in the case of using the redefined pattern, Wherein the step of obtaining the reference block operates based on the redefined prediction vector.
7. 7. The decoding method of claim 6, further comprising: a step of dividing the current block into more than two sub-blocks, Wherein in the step of redefining the prediction vector, compensation correction of the prediction vector is performed individually for each of the sub-blocks based on the prediction vector, In the step of obtaining the reference block, two or more sub-reference blocks are obtained based on the prediction vector redefined for each of the sub-blocks separately, and the reference block is obtained in combination with each of the sub-reference blocks.
8. 8. The decoding method according to claim 6 or 7, wherein, The step of redefining the prediction vector operates based on a template matching method having a search area referenced to a location indicated by the prediction vector.
9. 9. The decoding method according to claim 6 or 7, wherein, In the step of redefining the prediction vector, compensation correction is performed with a precision higher than that of the prediction vector.
10. 10. The decoding method of claim 9, wherein, In the step of redefining the prediction vector, the prediction vector compensation read in integer pixel units is corrected to sub-pixel units.
11. 11. The decoding method of claim 1, further comprising: A step of judging whether or not the prediction mode information indicates a mode allowing two or more prediction vectors, Wherein the step of reading the prediction vector comprises the step of reading a first prediction vector and the step of reading a second prediction vector, The step of obtaining the reference block includes the steps of obtaining a first reference block based on the first prediction vector and obtaining a second reference block based on the second prediction vector, The step of generating the prediction block generates a prediction block by prediction fusion based on two or more reference blocks including the first reference block and the second reference block.
12. 12. The decoding method of claim 11, wherein, The first prediction vector and the second prediction vector are block vectors for intra prediction, The first and second prediction vectors are configured to represent positions of different reference blocks in a decoded region of a current image.
13. 13. The decoding method of claim 11, wherein, The first prediction vector is a block vector for intra prediction and is configured to represent a position of the first reference block in a decoded region of a current image, The second prediction vector is a motion vector for inter-picture prediction and is configured to represent a position of the second reference block in a previously finished decoded image.
14. 14. The decoding method of claim 11, wherein, The prediction fusion is performed by a weighted sum or a weighted average performed by a weighted combination, The weighted information is obtained from the bit stream.
15. 15. An encoding method for generating video encoding into a bitstream, comprising: a step of determining at least one prediction vector for a current block of a current image being encoded, wherein the prediction vector is used to specify a position of at least one reference block from at least one image including the current image being decoded; a step of obtaining at least one reference block based on the prediction vector; a step of generating a prediction block based on the at least one reference block; A step of performing predictive coding of the current block based on the predicted block; a step of determining a prediction mode of the current block as an intra block copy IBC-based mode based on a result of the prediction encoding; generating a bitstream syntax representing the prediction mode and the at least one prediction vector based on the prediction coding mode, and And recording the bit stream grammar to the bit stream.
16. 16. The encoding method according to claim 15, wherein, The prediction vector is determined by at least one precision comprising sub-pixel units, The step of generating the bitstream syntax includes the step of generating a bitstream syntax for representing a unit of the precision based on the adaptive block vector resolution ABVR flag value and precision expression information of the block vector, and the step of generating a bitstream syntax for representing the prediction vector based on the unit of the precision.
17. 17. The encoding method according to claim 15, further comprising: a step of compensating and correcting the prediction vector, Wherein the step of obtaining the reference block operates based on the compensation corrected prediction vector, In the step of determining the prediction mode, the prediction mode is determined as a redefined mode using a prediction vector, In the generating of the bitstream syntax, a bitstream syntax representing the prediction vector before compensation correction is generated.
18. 18. The encoding method according to claim 15, wherein, The step of determining the prediction vector comprises the steps of determining a first prediction vector and determining a second prediction vector, The step of obtaining the reference block includes the steps of obtaining a first reference block based on the first prediction vector and obtaining a second reference block based on the second prediction vector, A step of generating the prediction block by generating a prediction block based on prediction fusion of two or more reference blocks including the first reference block and the second reference block, In the step of determining the prediction mode, the prediction mode is determined as a mode allowing two or more prediction vectors.
19. 19. The encoding method of claim 18, wherein, The first prediction vector is a block vector for intra prediction and is configured to represent a position of the first reference block in an encoded region of a current image, The second prediction vector is a motion vector for inter-picture prediction and is configured to represent a position of the second reference block in a previously encoded image.
20. 20. A decoder device configured to decode, by a computing device, an encoded video bitstream, comprising: A processor; a memory; an input section for inputting the bit stream; An output section for outputting the decoded video; A reference buffer storing at least one decoded image information including an image currently being decoded; A bit stream analysis unit that includes a function of reading prediction mode information from the bit stream, wherein the prediction mode information is used for predictive decoding of a current block of the image being currently decoded, a function of judging whether the prediction mode information indicates a mode of copying IBC based on an intra block, and a function of reading at least one prediction vector from the bit stream based on the prediction mode information, wherein the prediction vector specifies a position of at least one reference block from at least one of the images; a predictive decoding section including a function of obtaining at least one reference block from at least one image stored in the reference buffer based on the predictive vector, generating a predictive block based on the at least one reference block, and performing predictive decoding of the current block based on the predictive block, and And a video decoding section configured to decode the bitstream based on the predictive decoding result.

Description

Video encoding/decoding method and device based on intra-picture prediction Technical Field The present invention relates to the field of encoding (encoding) and decoding (decoding) of video (digital video), and to a digital video encoding and decoding method, a method for recording data thereof, and components, devices and systems for implementing the method. In particular, the present invention relates to an improved encoding/decoding method and apparatus based on Intra Block Copy (IBC) in an intra prediction encoding/decoding method. The present invention may be in the same technical field as at least one of the known digital video compression technical standards (e.g., MPEG-2, MPEG-4 Video, H.263, H.264/AVC, H.265/HEVC, H.266/VVC, VC-1, AV1, quickTime, VP-9, VP-10, and Motion JPEG), or may be a field for improving the intrinsic efficiency of such standards, or may be used to improve or replace such standards. Background Digital video encoding and decoding are widely used in various digital video application fields. For example, devices such as digital television broadcasting, video transmission through a communication network, video calls, video dialogues, video chats, video content recorded and played using optical discs including VCD (video compact disc)/DVD (digital versatile disc)/Blu-Ray disc (Blu-Ray), full processes for making, editing, collecting and distributing video content, and video photographing and recording actions including personal, business, industrial, and security monitoring purposes, etc., all rely on video encoding and decoding techniques. Accordingly, what may be referred to as digital video encoders and decoders may form part of a digital television, digital broadcast system, wireless broadcast system, notebook computer, desktop computer, tablet computer, electronic book reader, digital camera, digital video recording device, digital multimedia playback device, video game device/terminal/console, mobile phone (including smart phones) with multimedia playback capabilities, equipment for video conferencing, and other wide range of devices related to the generation, recording, and playback of digital video. The above-described digital video encoder and decoder may be implemented using digital video compression standards widely used and understood by those skilled in the art. The digital video compression standard may include at least one known compression standard, such as the standard names MPEG-2, MPEG-4 Video, H.263, H.264/AVC, H.265/HEVC, H.266/VVC, VC-1, AV1, quickTime, VP-9, VP-10, motion JPEG. Video encoders and decoders may more efficiently implement encoding or decoding of digital video information by meeting or modifying or deforming the above-described standards. Modification attempts to the above criteria may also lead to new criteria. In a well known case, there is an enhanced compression model (enhanced compression model, ECM) developed by the joint international standardization organization for ISO, IEC, ITU-T, joint video expert group (joint video experts team, JVET), which aims to improve and replace the existing h.266/VVC standard. A general video coding (VERSATILE VIDEO CODING, VVC), which is an international standard for video compression, is an intra prediction coding/decoding method that provides a method of deriving block information of a target block so that the same intra reference block is used as prediction information of the target block. This technique is considered a useful compression tool for screen content of computer graphics processing, collectively referred to as Intra Block Copy (IBC). Disclosure of Invention Problems to be solved by the invention It is well known that digital video, in an uncompressed state, requires a large amount of information to describe its content. Thus, the efficiency of recording or transmitting such information in its original form may be low. Therefore, digital video needs to be compressed in a variety of ways before it can be recorded or transmitted. The compression method includes lossy encoding and lossless encoding. Lossy encoding achieves high compression performance by sacrificing part of image quality, while lossless encoding prevents degradation of image quality by sacrificing part of compression performance. Regardless of the encoding method employed, in order to meet the demand for obtaining high-quality digital video within a limited storage recording capacity and communication transmission bandwidth, it is required to realize a technique capable of obtaining a high compression rate while minimizing the sacrifice of image quality. As described above, the encoding process for compression requires various operations such as spatial division of digital video, division and/or processing of color channels, removal of spatial redundancy (redundancy), removal of temporal redundancy, tracking of motion vectors (motion vectors) within video, encoding of differential images (DIFFERENTIAL IMAGE), quantization (quantiz