US 12,439,038 B2
Reduced complexity multi-mode neural network filtering of video data
Hongtao Wang, San Diego, CA (US); Samuel James Eadie, Munich (DE); Muhammed Zeyd Coban, Carlsbad, CA (US); and Marta Karczewicz, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Jul. 3, 2023, as Appl. No. 18/346,620.
Claims priority of provisional application 63/367,712, filed on Jul. 5, 2022.
Prior Publication US 2024/0015284 A1, Jan. 11, 2024
Int. Cl. H04N 11/02 (2006.01); H04N 19/117 (2014.01); H04N 19/176 (2014.01); H04N 19/80 (2014.01)
CPC H04N 19/117 (2014.11) [H04N 19/176 (2014.11); H04N 19/80 (2014.11)] 26 Claims
OG exemplary drawing
 
1. A method of filtering video data, the method comprising:
applying one or more neural network processing blocks to intermediate filtered video data, each of the neural network processing blocks including a first 1×1 convolutional filter, a parametric rectified linear unit (PReLU) filter, a second 1×1 convolutional filter, and a 3×3 convolutional filter, wherein the first 1×1 convolutional filter provides output to the PRELU filter, the PRELU filter provides output to the second 1×1 convolutional filter, and the second 1×1 convolutional filter provides output to the 3×convolutional filter;
applying additional neural network processing blocks to output of the one or more neural network processing blocks to form filtered video data; and
outputting the filtered video data.