US 12,143,573 B2
Neural network based coefficient sign prediction field
Xin Zhao, Palo Alto, CA (US); Yixin Du, Palo Alto, CA (US); Liang Zhao, Palo Alto, CA (US); Madhu Peringassery Krishnan, Palo Alto, CA (US); and Shan Liu, Palo Alto, CA (US)
Assigned to TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed by TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed on Sep. 11, 2023, as Appl. No. 18/244,768.
Application 18/244,768 is a continuation of application No. 17/494,238, filed on Oct. 5, 2021, granted, now 11,792,391.
Application 17/494,238 is a continuation of application No. 17/071,582, filed on Oct. 15, 2020, granted, now 11,190,760, issued on Nov. 30, 2021.
Prior Publication US 2023/0421754 A1, Dec. 28, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 19/105 (2014.01); G06N 3/08 (2023.01); H04N 19/129 (2014.01); H04N 19/132 (2014.01); H04N 19/176 (2014.01)
CPC H04N 19/105 (2014.11) [G06N 3/08 (2013.01); H04N 19/129 (2014.11); H04N 19/132 (2014.11); H04N 19/176 (2014.11)] 20 Claims
OG exemplary drawing
 
1. A method of video decoding, executable by a processor, the method comprising:
receiving a video bitstream comprising a current block in a current picture of video data of the video bitstream;
identifying reference samples and magnitudes of transform coefficients, corresponding to the current block, from an input to a neural network;
predicting sign values associated with the transform coefficients based at least on the identified reference samples, wherein the sign values are predicted using a convolutional neural network; and
decoding the video data based on the predicted sign values.