EP-4742671-A2 - METHODS FOR DECODING OR CODING PREDICTION MODE, DECODING OR CODING APPARATUS, AND STORAGE MEDIUM

Abstract

Provided is a method for decoding a prediction mode. In response to determining that a merge mode is enabled for a current block and a current candidate prediction mode is allowed for the current block, acquiring enablement states of one or more to-be-traversed prediction modes; decoding indication information of the current candidate prediction mode from a code stream in response to determining that a candidate prediction mode being allowed is present in to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block, and determining that the current candidate prediction mode is enabled for the current block without decoding the indication information of the current candidate prediction mode from the code stream in response to determining that none of the to-be-traversed prediction modes is allowed.

Inventors

• CHEN, Fangdong

Assignees

• Hangzhou Hikvision Digital Technology Co., Ltd.

Dates

Publication Date

20260513

Application Date

20200619

Claims (15)

1. A method for decoding a prediction mode, comprising: in response to determining that a merge mode is enabled for a current block and a current candidate prediction mode is allowed for the current block, executing the following steps: acquiring enablement states of one or more to-be-traversed prediction modes, wherein the one or more to-be-traversed prediction modes comprise at least one candidate prediction mode; decoding indication information of the current candidate prediction mode from a code stream in response to determining that a candidate prediction mode being allowed is present in to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block, and determining that the current candidate prediction mode is enabled for the current block without decoding the indication information of the current candidate prediction mode from the code stream in response to determining that none of the to-be-traversed prediction modes is allowed; wherein acquiring the enablement states of the to-be-traversed prediction modes comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the mode restriction condition comprises a size restriction condition, a frame type restriction condition, and a sequence level switch restriction condition, the enablement states of the to-be-traversed prediction modes refer to whether the candidate prediction mode is allowed for the current block in the to-be-traversed prediction modes, the to-be-traversed prediction modes refer to prediction modes, other than the current candidate prediction mode, of which the enablement state for the current block has not been determined yet; wherein the current candidate prediction mode and the at least one candidate prediction mode of the to-be-traversed prediction modes are all merge modes at least comprise a regular merge mode, a merge mode with motion vector difference, a sub-block motion information based merge mode, a geometric partitioning mode with triangular partition (triangle prediction mode, TPM), or a combined inter-picture merge and intra-picture prediction mode; the size restriction condition corresponding to the regular merge mode is that an area of the current block is greater than or equal to a first threshold, the frame type restriction condition corresponding to the regular merge mode is that the current block is a non-I frame; the size restriction condition corresponding to the merge mode with motion vector difference is that the area of the current block is greater than or equal to the first threshold, the frame type restriction condition corresponding to the merge mode with motion vector difference is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the merge mode with motion vector difference is that the sequence-level switch of the current block enables the merge mode with motion vector difference; the size restriction condition corresponding to the subblock-based merge mode is that both the width and height of the current block are greater than or equal to a second threshold, the frame type restriction condition corresponding to the subblock-based merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the subblock-based merge mode is that the sequence-level switch of the current block enables the affine merge mode or the SBTMVP mode; the size restriction condition corresponding to the combined inter and intra prediction merge mode is that the area of the current block is greater than or equal to a third threshold, and both the height and width are less than the size threshold CTU_SIZE, the frame type restriction condition corresponding to the combined inter and intra prediction merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the combined inter and intra prediction merge mode is that the sequence-level switch of the current block enables the combined inter and intra prediction merge mode; the size restriction condition corresponding to the triangular prediction merge mode is that the area of the current block is greater than or equal to the third threshold, the frame type restriction condition corresponding to the triangular prediction merge mode is that the current block is a B frame, and the sequence-level switch restriction condition corresponding to the triangular prediction merge mode is that the sequence-level switch of the current block enables the triangular prediction merge mode.
2. The method according to claim 1, wherein acquiring the enablement states of the to-be-traversed prediction modes comprises: acquiring an enablement state of each candidate prediction mode in the to-be-traversed prediction modes according to a traversal sequence of the to-be-traversed prediction modes, wherein the traversal sequence refers to a determining order for sequentially determining whether each candidate prediction mode in the to-be-traversed prediction modes is allowed, and the current candidate prediction mode precedes the to-be-traversed prediction modes; determining that none of the to-be-traversed prediction modes is allowed in response to traversing to a last candidate prediction mode in the traversal sequence that is still not allowed, or, determining that a candidate prediction mode allowed is present in the to-be-traversed prediction modes in response to traversing to a candidate prediction mode that is allowed.
3. The method according to claim 1, further comprising: directly setting a value of a mode flag value of the current candidate prediction mode to 0 without decoding the mode flag value of the current candidate prediction mode, in response to determining that the current candidate prediction mode is not allowed for the current block.
4. The method according to claim 1, wherein the current candidate prediction mode is the regular merge mode; the to-be-traversed prediction modes comprise the geometric partitioning mode with triangular partition, the merge mode with motion vector difference, the sub-block motion information based merge mode, and the combined inter-picture merge and intra-picture prediction mode; and determining that none of the to-be-traversed prediction modes is allowed comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that for each candidate prediction mode in the to-be-traversed prediction modes, a sequence level enabled flag for each candidate prediction mode is in a disabled state.
5. The method according to claim 1, wherein: the current candidate prediction mode is the merge mode with motion vector difference, and the to-be-traversed prediction modes include the sub-block merge mode, the combined inter-picture merge and intra-picture prediction mode, and the geometric partitioning mode with triangular partition; and determining that none of the to-be-traversed prediction modes is allowed comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the area of the current block is smaller than the area threshold, wherein the area threshold is determined based on the size restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the second threshold is 8, the third threshold is determined by a threshold of height and/or width, the threshold of height and/or width is 8.
6. The method according to claim 1, wherein: the current candidate prediction mode is the sub-block merge mode, and the to-be-traversed prediction modes include the combined inter-picture merge and intra-picture prediction mode and the geometric partitioning mode with triangular partition; and determining that none of the to-be-traversed prediction modes is allowed comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of the geometric partitioning mode with triangular partition but meets a specified condition, where the specified condition includes one or more of the following conditions: the sequence level enabled flag for the combined inter-picture merge and intra-picture prediction mode in the current block is in a disabled state; the prediction mode of the current block belongs to a skip mode, wherein the third threshold is determined by a threshold of height and/or width, the threshold of height and/or width is 8.
7. The method according to claim 1, wherein: the current candidate prediction mode is the combined inter-picture merge and intra-picture prediction mode, and the to-be-traversed prediction modes include the geometric partitioning mode with triangular partition; and determining that none of the to-be-traversed prediction modes is allowed comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet the mode restriction condition of the geometric partitioning mode with triangular partition, wherein the third threshold is determined by a threshold of height and/or width, the threshold of height and/or width is 8.
8. The method according to claim 1, further comprising: determining that the current candidate prediction mode is not used for the current block in response to determining that the current candidate prediction mode is not allowed for the current block.
9. The method according to claim 1, wherein acquiring the enablement states of the one or more to-be-traversed prediction modes comprises: determining the combined inter-picture merge and intra-picture prediction mode is not allowed for the current block in response to the prediction mode of the current block belonging to a skip mode, or, determining the combined inter-picture merge and intra-picture prediction mode is not allowed for the current block in response to a height and a width of the current block being not both less than 128; optionally, wherein acquiring the enablement states of the one or more to-be-traversed prediction modes comprises: determining the merge mode with motion vector difference is not allowed for the current block in response to the sequence level enabled flag for the merge mode with motion vector difference in the current block being 0, wherein a value of the sequence level enabled flag is stored in a sequence parameter set; optionally, wherein acquiring the enablement states of the one or more to-be-traversed prediction modes comprises: determining the sub-block motion information based merge mode is not allowed for the current block in response to a size of the current block being 4×8 or 8×4, optionally the current block not meeting the mode restriction condition of the geometric partitioning mode with triangular partition comprises: the image frame where the current block is located is a P frame, optionally, wherein the size threshold CTU_SIZE is 128.
10. A method for decoding a prediction mode, comprising: in response to determining that a merge mode is enabled for a current block and a current candidate prediction mode is allowed for the current block, executing the following steps: acquiring enablement states of one or more to-be-traversed prediction modes, wherein the one or more to-be-traversed prediction modes comprise at least one candidate prediction mode; decoding indication information of the current candidate prediction mode from a code stream in response to determining that a candidate prediction mode being allowed is present in to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block, and determining that the current candidate prediction mode is enabled for the current block without decoding the indication information of the current candidate prediction mode from the code stream in response to determining that none of the to-be-traversed prediction modes is allowed; wherein acquiring the enablement states of the to-be-traversed prediction modes comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the mode restriction condition comprises a sequence level switch restriction condition, the enablement states of the to-be-traversed prediction modes refer to whether the candidate prediction mode is allowed for the current block in the to-be-traversed prediction modes, the to-be-traversed prediction modes refer to prediction modes, other than the current candidate prediction mode, of which the enablement state for the current block has not been determined yet; wherein the current candidate prediction mode and the at least one candidate prediction mode of the to-be-traversed prediction modes are all merge modes at least comprise a regular merge mode, a merge mode with motion vector difference, a sub-block motion information based merge mode, a geometric partitioning mode with triangular partition (triangle prediction mode, TPM), or a combined inter-picture merge and intra-picture prediction mode; wherein determining that none of the to-be-traversed prediction modes is allowed comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that for each candidate prediction mode in the to-be-traversed prediction modes, a sequence level enabled flag for each candidate prediction mode is in a disabled state.
11. A method for coding a prediction mode, comprising: in response to determining that a merge mode is enabled for a current block and a current candidate prediction mode is allowed for the current block, executing the following steps: acquiring enablement states of one or more to-be-traversed prediction modes, wherein the one or more to-be-traversed prediction modes comprise at least one candidate prediction mode allowed for predicting a current block; coding indication information of the current candidate prediction mode into a code stream in response to determining that a candidate prediction mode being allowed is present in the to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block; and setting that the current candidate prediction mode is enabled for the current block and skip encoding the indication information of the current candidate prediction mode into the code stream in response to determining that none of the to-be-traversed prediction modes is allowed; wherein the current candidate prediction mode and the at least one candidate prediction mode of the one or more to-be-traversed prediction modes are all merge modes at least comprising a regular merge mode, a merge mode with motion vector difference, a sub-block motion information based merge mode, a geometric partitioning mode with triangular partition (triangle prediction mode, TPM), or a combined inter-picture merge and intra-picture prediction mode; wherein acquiring the enablement states of the to-be-traversed prediction modes comprises: determining that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the mode restriction condition comprises a size restriction condition, a restriction condition of other modes, a frame type restriction condition, and a sequence level switch restriction condition, the enablement states of the to-be-traversed prediction modes refer to whether the candidate prediction mode is allowed for the current block in the to-be-traversed prediction modes, the to-be-traversed prediction modes refer to prediction modes, other than the current candidate prediction mode, of which the enablement condition for the current block has not been determined yet; the size restriction condition corresponding to the regular merge mode is that an area of the current block is greater than or equal to a first threshold, the frame type restriction condition corresponding to the regular merge mode is that the current block is a non-I frame; the size restriction condition corresponding to the merge mode with motion vector difference is that the area of the current block is greater than or equal to the first threshold, the frame type restriction condition corresponding to the merge mode with motion vector difference is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the merge mode with motion vector difference is that the sequence-level switch of the current block enables the merge mode with motion vector difference; the size restriction condition corresponding to the subblock-based merge mode is that both the width and height of the current block are greater than or equal to a second threshold, the frame type restriction condition corresponding to the subblock-based merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the subblock-based merge mode is that the sequence-level switch of the current block enables the affine merge mode or the SBTMVP mode; the size restriction condition corresponding to the combined inter and intra prediction merge mode is that the area of the current block is greater than or equal to a third threshold, and both the height and width are less than the size threshold CTU_SIZE, the frame type restriction condition corresponding to the combined inter and intra prediction merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the combined inter and intra prediction merge mode is that the sequence-level switch of the current block enables the combined inter and intra prediction merge mode; the size restriction condition corresponding to the triangular prediction merge mode is that the area of the current block is greater than or equal to the third threshold, the frame type restriction condition corresponding to the triangular prediction merge mode is that the current block is a B frame, and the sequence-level switch restriction condition corresponding to the triangular prediction merge mode is that the sequence-level switch of the current block enables the triangular prediction merge mode.
12. A decoding apparatus, comprising: an acquiring module, configured to execute the following steps in response to determining that a merge mode is enabled for the current block and the current candidate prediction mode is allowed for the current block: acquire enablement states of one or more to-be-traversed prediction modes, wherein the one or more to-be-traversed prediction modes comprise at least one candidate prediction mode; a determining module, configured to decode indication information of the current candidate prediction mode from a code stream in response to determining that a candidate prediction mode being allowed is present in to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block, and determining that the current candidate prediction mode is enabled for the current block without decoding the indication information of the current candidate prediction mode from the code stream in response to determining that none of the to-be-traversed prediction modes is allowed; wherein the acquiring module is further configured to determine that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the mode restriction condition comprises a size restriction condition, a frame type restriction condition, and a sequence level switch restriction condition, the enablement states of the to-be-traversed prediction modes refer to whether the candidate prediction mode is allowed for the current block in the to-be-traversed prediction modes, the to-be-traversed prediction modes refer to prediction modes, other than the current candidate prediction mode, of which the enablement state for the current block has not been determined yet; wherein the current candidate prediction mode and the at least one candidate prediction mode of the to-be-traversed prediction modes are all merge modes at least comprise a regular merge mode, a merge mode with motion vector difference, a sub-block motion information based merge mode, a geometric partitioning mode with triangular partition (triangle prediction mode, TPM), or a combined inter-picture merge and intra-picture prediction mode; the size restriction condition corresponding to the regular merge mode is that an area of the current block is greater than or equal to a first threshold, the frame type restriction condition corresponding to the regular merge mode is that the current block is a non-I frame; the size restriction condition corresponding to the merge mode with motion vector difference is that the area of the current block is greater than or equal to the first threshold, the frame type restriction condition corresponding to the merge mode with motion vector difference is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the merge mode with motion vector difference is that the sequence-level switch of the current block enables the merge mode with motion vector difference; the size restriction condition corresponding to the subblock-based merge mode is that both the width and height of the current block are greater than or equal to a second threshold, the frame type restriction condition corresponding to the subblock-based merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the subblock-based merge mode is that the sequence-level switch of the current block enables the affine merge mode or the SBTMVP mode; the size restriction condition corresponding to the combined inter and intra prediction merge mode is that the area of the current block is greater than or equal to a third threshold, and both the height and width are less than the size threshold CTU_SIZE, the frame type restriction condition corresponding to the combined inter and intra prediction merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the combined inter and intra prediction merge mode is that the sequence-level switch of the current block enables the combined inter and intra prediction merge mode; the size restriction condition corresponding to the triangular prediction merge mode is that the area of the current block is greater than or equal to the third threshold, the frame type restriction condition corresponding to the triangular prediction merge mode is that the current block is a B frame, and the sequence-level switch restriction condition corresponding to the triangular prediction merge mode is that the sequence-level switch of the current block enables the triangular prediction merge mode.
13. A coding apparatus, comprising: an acquiring module, configured to acquire enablement states of one or more to-be-traversed prediction modes in response to determining that a merge mode is enabled for a current block and a current candidate prediction mode is allowed for the current block, wherein the one or more to-be-traversed prediction modes comprise at least one candidate prediction mode allowed for predicting a current block other than a current candidate prediction mode; a coding module, configured to code indication information of the current candidate prediction mode into a code stream in response to determining that a candidate prediction mode being allowed is present in the to-be-traversed prediction modes, wherein the indication information is configured to indicate whether to enable the current candidate prediction mode for the current block; and set that the current candidate prediction mode is enabled for the current block and skip encoding the indication information of the current candidate prediction mode into the code stream in response to determining that none of the to-be-traversed prediction modes is allowed; wherein the current candidate prediction mode and the at least one candidate prediction mode of the one or more to-be-traversed prediction modes are all merge modes at least comprising a regular merge mode, a merge mode with motion vector difference, a sub-block motion information based merge mode, a geometric partitioning mode with triangular partition (triangle prediction mode, TPM), or a combined inter-picture merge and intra-picture prediction mode; wherein the acquiring module is further configured to determine that none of the to-be-traversed prediction modes is allowed in response to determining that the current block does not meet a mode restriction condition of each candidate prediction mode in the to-be-traversed prediction modes, wherein the mode restriction condition comprises a size restriction condition, a restriction condition of other modes, a frame type restriction condition, and a sequence level switch restriction condition, the enablement states of the to-be-traversed prediction modes refer to whether the candidate prediction mode is allowed for the current block in the to-be-traversed prediction modes, the to-be-traversed prediction modes refer to prediction modes, other than the current candidate prediction mode, of which the enablement condition for the current block has not been determined yet; wherein the size restriction condition corresponding to the regular merge mode is that an area of the current block is greater than or equal to a first threshold, the frame type restriction condition corresponding to the regular merge mode is that the current block is a non-I frame; the size restriction condition corresponding to the merge mode with motion vector difference is that the area of the current block is greater than or equal to the first threshold, the frame type restriction condition corresponding to the merge mode with motion vector difference is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the merge mode with motion vector difference is that the sequence-level switch of the current block enables the merge mode with motion vector difference; the size restriction condition corresponding to the subblock-based merge mode is that both the width and height of the current block are greater than or equal to a second threshold, the frame type restriction condition corresponding to the subblock-based merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the subblock-based merge mode is that the sequence-level switch of the current block enables the affine merge mode or the SBTMVP mode; the size restriction condition corresponding to the combined inter and intra prediction merge mode is that the area of the current block is greater than or equal to a third threshold, and both the height and width are less than the size threshold CTU_SIZE, the frame type restriction condition corresponding to the combined inter and intra prediction merge mode is that the current block is a non-I frame, and the sequence-level switch restriction condition corresponding to the combined inter and intra prediction merge mode is that the sequence-level switch of the current block enables the combined inter and intra prediction merge mode; the size restriction condition corresponding to the triangular prediction merge mode is that the area of the current block is greater than or equal to the third threshold, the frame type restriction condition corresponding to the triangular prediction merge mode is that the current block is a B frame, and the sequence-level switch restriction condition corresponding to the triangular prediction merge mode is that the sequence-level switch of the current block enables the triangular prediction merge mode.
14. A decoding apparatus, comprising: a processor; and a memory configured to store instructions executable by the processor; wherein the processor is configured to execute the steps of the method according to any one of claims 1 to 9.
15. A coding apparatus, comprising: a processor; and a memory configured to store instructions executable by the processor; wherein the processor is configured to execute the steps of the method according to claim 11.

Description

TECHNICAL FIELD The present disclosure relates to the field of multimedia technologies, and in particular to methods for decoding or coding a prediction mode, a decoding or coding apparatus, and a storage medium. BACKGROUND The prediction mode refers to a mode for decoding or coding an image block in a video frame. As there are a plurality of prediction modes, for each prediction mode, one piece of indication information is configured to indicate whether the corresponding prediction mode is enabled. By coding the indication information at a coder-side, which prediction mode is enabled for the image block is indicated. A decoder-side may decode the coded indication information to determine the prediction mode enabled for the image block. SUMMARY Embodiments of the present disclosure provide methods for decoding or coding a prediction mode, a decoding or coding apparatus, and a storage medium. The technical solutions are as follows. According to a first aspect of embodiments of the present disclosure, a method for decoding a prediction mode according to claims 1-9 is provided. According to a second aspect of embodiments of the present disclosure, a method for decoding a prediction mode according to claim 10 is provided. According to a third aspect of embodiments of the present disclosure, a method for coding a prediction mode according to claims 11 is provided. According to a fourth aspect of embodiments of the present disclosure, a decoding apparatus according to claims 12 is provided. According to a fifth aspect of embodiments of the present disclosure, a coding apparatus according to claims 13 is provided. According to a sixth aspect of embodiments of the present disclosure, a decoding apparatus according to claims 14 is provided. According to a seventh aspect of embodiments of the present disclosure, a coding apparatus according to claims 15 is provided. The beneficial effects brought about by the technical solutions according to the embodiments of the present disclosure are as follows. When determining which prediction mode is enabled for the current block, for the current candidate prediction mode, the enablement states of the to-be-traversed prediction modes are acquired. The to-be-traversed prediction modes include at least one candidate prediction mode used for predicting the current block other than the current candidate prediction mode. Then, based on the enablement states of the to-be-traversed prediction modes, it can be determined whether to decode the indication information of the current candidate prediction mode. If none of the to-be-traversed prediction modes is allowed, the step of decoding the indication information can be skipped, and it can be directly determined that the current candidate prediction modem is used for the current block. Unlike the related art where the indication information of the current candidate prediction mode must be decoded to determine whether the prediction mode is enabled for the current block, the process of decoding the prediction mode is simplified and the efficiency in decoding the prediction mode is improved. BRIEF DESCRIPTION OF THE DRAWINGS To describe the technical solutions in the embodiments of the present disclosure more clearly, the following briefly introduces the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts. FIG. 1 is a flowchart of a method for decoding a prediction mode according to an embodiment of the present disclosure;FIG. 2 is a flowchart of a method for coding a prediction mode according to an embodiment of the present disclosure;FIG. 3 is a structural diagram of a decoding apparatus according to an embodiment of the present disclosure;FIG. 4 is a structural diagram of a coding apparatus according to an embodiment of the present disclosure; andFIG. 5 is a structural diagram of a coding and decoding apparatus according to an embodiment of the present disclosure. DETAILED DESCRIPTION For clearer description of the objectives, technical solutions, and advantages in the present disclosure, the embodiments of the present disclosure are described in further detail hereinafter with reference to the accompanying drawings. In the related art, when coding the current block, for any of the prediction modes, such as a prediction mode A, the coder-side judges whether the prediction mode is enabled. If that the prediction mode is used, the indication information of the prediction mode A is coded as 1; and if the prediction mode A is not used, then the indication information of the prediction mode A is coded as 0. In this way, the indication information of each prediction mode is coded. Subsequently, the decoder-side decodes the indication information of the prediction modes when the cu