Granted Patent B2

US 12,604,039 · App. 18/743,673 · Granted Apr 14, 2026

Inventors: Xiaoyu Xiu (San Diego, CA); Ning Yan (San Diego, CA); Yi-Wen Chen (San Diego, CA); Che-Wei Kuo (San Diego, CA); Wei Chen (San Diego, CA); Hong-Jheng Jhu (San Diego, CA); Han Gao (San Diego, CA); Xianglin Wang (San Diego, CA); Bing Yu (Beijing, CN)

Assignee: BEIJING DAJIA INTERNET INFORMATION TECHNOLOGY CO., LTD.

H04N19/60H04N19/124H04N19/18H04N19/91

Abstract

Claims (52)

1 . A video decoding method for sign prediction of transform coefficients, comprising:

receiving a bitstream comprising a sequence of sign signaling bits for a set of candidate transform coefficients;

generating a set of predicted signs for the set of candidate transform coefficients associated with a transform block of a video frame;

decoding the sequence of sign signaling bits based on one or more contexts used to entropy-encode the sequence of sign signaling bits to obtain an indication of correctness of the predicted signs of the respective candidate transform coefficients, wherein a context of the one or more contexts used to entropy-encode a sign signaling bit is determined based on a scan position of a candidate transform coefficient corresponding to the sign signaling bit; and

estimating original signs for the set of candidate transform coefficients based on the set of predicted signs and the sequence of decoded sign signaling bits.

2 . The video decoding method of claim 1 , wherein estimating the original signs for the set of candidate transform coefficients based on the set of predicted signs and the sequence of sign signaling bits comprises:

updating the predicted signs of the candidate transform coefficients according to the correctness indicated by the respective sign signaling bits to determine the original signs of the candidate transform coefficients.

3 . The video decoding method of claim 1 , wherein decoding the sequence of sign signaling bits comprises:

determining a context used to entropy-encode the sign signaling bit based on a magnitude of the candidate transform coefficient corresponding to the sign signaling bit.

4 . The video decoding method of claim 3 , wherein magnitudes of the candidate transform coefficients fall within a plurality of magnitude segments, wherein a plurality of contexts are assigned to entropy-encode sign signaling bits for transform coefficients belonging to the respective magnitude segments,

wherein determining the context used to entropy-encode the sign signaling bit comprises:

determining a magnitude segment, among the plurality of magnitude segments, to which the magnitude of the candidate transform coefficient corresponding to the sign signaling bit belongs; and

determining the context assigned to the magnitude segment as the context used to entropy-encode the sign signaling bit.

5 . The video decoding method of claim 4 , wherein every two consecutive magnitude segments, among the plurality of magnitude segments, are separated by a threshold.

6 . The video decoding method of claim 3 , wherein the bitstream further comprises encoded data of quantized levels of the set of candidate transform coefficients,

wherein the video decoding method further comprises:

determining the magnitude of the candidate transform coefficient by directly parsing a quantized level of the candidate transform coefficient from the bitstream without dequantization.

7 . The video decoding method of claim 3 , wherein the bitstream further comprises encoded data of quantized levels of the set of candidate transform coefficients,

wherein the video decoding method further comprises:

parsing a quantized level of the candidate transform coefficient from the bitstream;

determining a quantization index of the candidate transform coefficient based on the quantized level of the candidate transform coefficient and a transition state of the candidate transform coefficient; and

determining the magnitude of the candidate transform coefficient based on the quantization index of the candidate transform coefficient.

8 . The video decoding method of claim 1 , wherein the set of candidate transform coefficients are selected from transform coefficients of the transform block that are reordered according to their magnitudes,

wherein the video decoding method further comprises:

reordering the set of candidate transform coefficients parsed from the bitstream.

9 . The video decoding method of claim 8 , wherein the contexts used to entropy-encode the sign signaling bits are determined based on the magnitudes of reordered transform coefficients.

10 . The video decoding method of claim 8 , wherein the contexts used to entropy-encode the sign signaling bits are determined based on the magnitudes of transform coefficients before the reordering.

11 . The video decoding method of claim 1 , wherein scan positions of the transform coefficients are classified into a plurality of groups, wherein a plurality of contexts are assigned to entropy-encode sign signaling bits for transform coefficients belonging to the respective groups,

wherein the context of the one or more contexts used to entropy-encode a sign signaling bit is determined by:

determining a group, among the plurality of groups, to which the scan position of the candidate transform coefficient corresponding to the sign signaling bit belongs; and

determining the context assigned to the group as the context used to entropy-encode the sign signaling bit.

12 . The video decoding method of claim 1 , wherein decoding the sequence of sign signaling bits comprises:

determining a context used to entropy-encode the sign signaling bit based on coding mode, a block size or component channel information of the candidate transform coefficient corresponding to the sign signaling bit.

13 . The video decoding method of claim 1 , wherein the set of transform coefficients in the bitstream are generated by applying a plurality of different transform cores,

wherein decoding the sequence of sign signaling bits comprises:

determining a context used to entropy-encode the sign signaling bit based on the transform core applied to generate the candidate transform coefficient corresponding to the sign signaling bit.

14 . The video decoding method of claim 13 , wherein the transform cores comprise a multiple transform selection (MTS) and a low-frequency non-separable transform (LFNST).

15 . The video decoding method of claim 1 , wherein generating the set of predicted signs for the set of candidate transform coefficients comprises:

generating a plurality of candidate hypotheses for the set of candidate transform coefficients; and

selecting a hypothesis from the plurality of candidate hypotheses as the set of predicted signs for the set of candidate transform coefficients.

16 . A video decoding apparatus, comprising:

a memory configured to store a bitstream comprising a set of candidate transform coefficients and a sequence of sign signaling bits for the set of candidate transform coefficients; and

a processor coupled to the memory and configured to perform operations comprising:

receiving a bitstream comprising a sequence of sign signaling bits for a set of candidate transform coefficients;

generating a set of predicted signs for the set of candidate transform coefficients associated with a transform block of a video frame;

decoding the sequence of sign signaling bits based on one or more contexts used to entropy-encode the sequence of sign signaling bits to obtain an indication of correctness of the predicted signs of the respective candidate transform coefficients, wherein a context of the one or more contexts used to entropy-encode a sign signaling bit is determined based on a scan position of a candidate transform coefficient corresponding to the sign signaling bit; and

estimating original signs for the set of candidate transform coefficients based on the set of predicted signs and the sequence of decoded sign signaling bits.

17 . A non-transitory computer-readable storage medium having stored therein a bitstream comprising a sequence of sign signaling bits for a set of candidate transform coefficients, wherein the bitstream is decodable by performing operations comprising:

receiving a bitstream comprising a sequence of sign signaling bits for a set of candidate transform coefficients;

generating a set of predicted signs for the set of candidate transform coefficients associated with a transform block of a video frame;

decoding the sequence of sign signaling bits based on one or more contexts used to entropy-encode the sequence of sign signaling bits to obtain an indication of correctness of the predicted signs of the respective candidate transform coefficients, wherein a context of the one or more contexts used to entropy-encode a sign signaling bit is determined based on a scan position of a candidate transform coefficient corresponding to the sign signaling bit; and

estimating original signs for the set of candidate transform coefficients based on the set of predicted signs and the sequence of decoded sign signaling bits.

Assignments (4)

Continuity (13)

References Cited (50)