US 11,863,783 B2
Artificial intelligence-based image encoding and decoding apparatus and method
Quockhanh Dinh, Suwon-si (KR); Minwoo Park, Suwon-si (KR); Minsoo Park, Suwon-si (KR); and Kwangpyo Choi, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Feb. 22, 2022, as Appl. No. 17/677,498.
Application 17/677,498 is a continuation of application No. PCT/KR2022/002493, filed on Feb. 21, 2022.
Claims priority of application No. 10-2021-0023695 (KR), filed on Feb. 22, 2021; application No. 10-2021-0123369 (KR), filed on Sep. 15, 2021; and application No. 10-2021-0171269 (KR), filed on Dec. 2, 2021.
Prior Publication US 2022/0272372 A1, Aug. 25, 2022
Int. Cl. H04N 19/51 (2014.01); H04N 19/137 (2014.01); H04N 19/124 (2014.01); H04N 19/43 (2014.01); H04N 19/91 (2014.01); H04N 19/17 (2014.01)
CPC H04N 19/51 (2014.11) [H04N 19/124 (2014.11); H04N 19/137 (2014.11); H04N 19/17 (2014.11); H04N 19/43 (2014.11); H04N 19/91 (2014.11)] 11 Claims
OG exemplary drawing
 
1. A method of reconstructing an optical flow by using artificial intelligence (AI), the method comprising:
obtaining, from a bitstream, feature data of a current residual optical flow for a current image;
obtaining the current residual optical flow by applying the feature data of the current residual optical flow to a neural-network-based first decoder;
obtaining a second-order optical flow between a current predicted optical flow and a previous optical flow; and
obtaining the current predicted optical flow by modifying the previous optical flow according to the second-order optical flow; and
reconstructing a current optical flow based on the current residual optical flow and the current predicted optical flow.