ERROR RESILIENCE AND PARALLEL PROCESSING FOR DECODER SIDE MOTION VECTOR DERIVATION

• US 20200137413A1
• Filed: 12/30/2019
• Published: 04/30/2020
• Est. Priority Date: 06/30/2017
• Status: Abandoned Application

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: ESENLIK, Semih, ZHAO, Zhijie, KOTRA, Anand Meher
• Application Number:
  US16/730,825
• Publication Number:
  US 20200137413A1
• Time in Patent Office: Days
• CPC Class Codes: H04N 19/436 using parallelised computational arrangements
• CPC Class Codes: H04N 19/44 Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
• CPC Class Codes: H04N 19/52 by predictive encoding
• CPC Class Codes: H04N 19/521 for estimating the reliability of the determined motion vectors or motion vector field, e.g. for smoothing the motion vector field or for correcting motion vectors
• CPC Class Codes: H04N 19/56 Motion estimation with initialisation of the vector search, e.g. estimating a good candidate to initiate a search
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques

Abstract

The present disclosure relates to inter-prediction in which it is determined whether or not a motion vector of a block neighbor to the current block has been determined by the template matching, as used for instance in decoder-based motion vector derivation and/or refinement. If detected negatively, prediction of the current block is derived according to the motion vector of the neighbor block. On the other hand, if detected positively, the prediction of the current block is not based on the neighbor block.

Claims

1. An apparatus for determining a prediction of a current block, for reconstruction of the current block, the apparatus comprising:

a prediction unit configured to detect whether or not a motion vector of a block adjacent to the current block has been determined by template matching, and configured to;

if detected negatively, derive prediction of the current block according to the adjacent block,

if detected positively, not to derive prediction of the current block according to the adjacent block.

2. The apparatus according to claim 1, configured for determining a motion vector to determine inter-prediction of a current block, for reconstruction of the current block, the prediction unit comprising:

a motion vector candidate list determination unit configured to form a list of candidate motion vectors for the current block by;

detecting whether or not a motion vector of an adjacent block has been determined by the template matching,

if detected negatively, including the motion vector of the adjacent block as a candidate into the list,

if detected positively, not including the motion vector of the adjacent block as a candidate into the list, and

a motion vector determination unit configured to select an estimate of a motion vector from the list and determining the motion vector of the current block according to the estimate.

3. The apparatus according to claim 2, wherein the motion vector candidate list determination unit is configured to, if detected positively, include a replacement motion vector as a candidate in the list.

4. The apparatus according to claim 3, wherein the replacement motion vector is a motion vector of a block neighbor to the current block and different from the adjacent block.

5. The apparatus according to claim 2, further comprising

a motion vector refinement unit configured to apply a refinement of the selected estimate of the motion vector,

the motion vector candidate list determination unit is configured to, if detected positively, include into the list an estimate of the motion vector of the adjacent block obtained by the template matching before applying the refinement,

the motion vector determination unit is further configured to determine the motion vector of the adjacent block according to the estimate of the motion vector of the adjacent block after the refinement.

6. The apparatus according to claim 5, wherein the motion vector candidate list determination unit is configured to form the list for the current block while the motion vector refinement unit applies refinement to the adjacent block.

7. The apparatus according to claim 5, wherein

the motion vector candidate list determination unit is configured to, if detected positively,

if the adjacent block is located in the same picture as the current block, include in the list an estimate of the motion vector of the adjacent block obtained by the template matching before applying the refinement, and

if the adjacent block is located in a picture other than the current block, include in the list the motion vector of the adjacent block after applying the refinement to the adjacent block.

8. The apparatus according to claim 2, wherein

the motion vector candidate list determination unit is further configured to check whether a motion vector to be included into the list is already included in the list, and not including it into the list if affirmative, and

said checking is not performed when including in the list a motion vector of an adjacent block to be processed at least partially in parallel with the processing of the current block.

9. The apparatus according to claim 1, for determining intra-prediction of a current block, for reconstruction of the current block, the prediction unit comprising:

a reference sample determination unit for determining a set of reference samples for the current block including;

detecting whether or not a motion vector of an adjacent block has been determined by the template matching,

if detected negatively, including one or more samples of the adjacent block into the set of reference samples,

if detected positively, not including any samples of the adjacent block into the set of reference samples, wherein

an intra-prediction unit for providing the prediction of the current block according to samples from the set of reference samples.

10. The apparatus according to claim 1, wherein the prediction unit is configured to, for reconstruction of the current block:

detect whether or not a motion vector of a surrounding block, which is located within a predetermined rectangular region defined with respect to the position of the current block, has been determined by the template matching.

11. A video decoder for decoding a plurality of pictures from a bitstream, comprising:

a bitstream parser for parsing the bitstream including control information, the apparatus according to claim 1, and

a reconstruction unit for reconstructing the current block based on the prediction of the current block.

12. The video decoder according to claim 11, wherein the control information includes:

a flag indicating whether or not the prediction unit is to detect whether or not a motion vector of an adjacent block has been determined by the template matching, and/or.

a replacement indicator indicating, if detected positively, how the prediction for the present block is determined.

13. The video decoder according to claim 12, wherein the control information is included in a sequence parameter set and/or in a slice header.

14. A video encoder for encoding a plurality of pictures into a bitstream, comprising:

the apparatus according to claim 1,

a bitstream former for including into the bitstream control information,

a reconstruction unit for reconstructing the current block based on the prediction of the current block and storing the reconstructed block in a memory.

15. The video encoder according to claim 14, wherein the control information includes:

a flag indicating whether or not the prediction unit is to detect whether or not a motion vector of an adjacent block has been determined by the template matching, and/or.

a replacement indicator indicating, if detected positively, how the prediction for the present block is determined.

16. The video encoder or decoder according to claim 15, wherein the control information is included in a sequence parameter set and/or in a slice header.

17. A method for determining a prediction of a current block, comprising:

detecting whether or not a motion vector of an adjacent block has been determined by the template matching,

if detected negatively, deriving prediction of the current block according to the adjacent block,

if detected positively, not deriving prediction of the current block according to the adjacent block.