US 11,936,912 B2
Method and apparatus for temporal smoothing for video
Jun Tian, Belle Mead, NJ (US); Wen Gao, West Windsor, NJ (US); and Shan Liu, San Jose, CA (US)
Assigned to TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed by Tencent America LLC, Palo Alto, CA (US)
Filed on Aug. 25, 2021, as Appl. No. 17/445,849.
Application 17/445,849 is a continuation of application No. 17/095,602, filed on Nov. 11, 2020, granted, now 11,140,416.
Claims priority of provisional application 63/000,292, filed on Mar. 26, 2020.
Prior Publication US 2021/0385495 A1, Dec. 9, 2021
Int. Cl. H04N 19/66 (2014.01); H04N 19/119 (2014.01); H04N 19/176 (2014.01); H04N 19/80 (2014.01); G06T 5/00 (2006.01)
CPC H04N 19/66 (2014.11) [H04N 19/119 (2014.11); H04N 19/176 (2014.11); H04N 19/80 (2014.11); G06T 5/002 (2013.01); G06T 5/005 (2013.01); G06T 2207/20182 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for video processing, comprising:
determining, by a processor, a frame interval for a current block in a current frame within a sequence of frames, the frame interval indicating a group of frames with collocated blocks of the current block that satisfy an error metric requirement comparing to the current block, a number of frames in the frame interval being determined based on a direct error threshold, a cumulative error threshold, and differences between the collocated blocks in the group of frames and the current block, the collocated block of each of the group of frames in the sequence of frames having a same location in the respective frame as the current block in the current frame;
determining, by the processor, a replacement block based on the collocated blocks in the group of frames; and
replacing, by the processor, the current block in the current frame with the replacement block,
wherein an error corresponding to the difference between each of the collocated blocks and the current block is less than the direct error threshold, and a sum of the errors corresponding to the differences between the collocated blocks and the current block is less than the cumulative error threshold.