US 12,425,615 B2
Methods, systems, and apparatuses for content-adaptive multi-layer coding based on neural networks
Dan Grois, Beer-Sheva (IL); and Alexander Giladi, Denver, CO (US)
Assigned to Comcast Cable Communications, LLC, Philadelphia, PA (US)
Filed by Comcast Cable Communications, LLC, Philadelphia, PA (US)
Filed on Oct. 8, 2021, as Appl. No. 17/497,588.
Prior Publication US 2023/0114562 A1, Apr. 13, 2023
Int. Cl. H04N 19/30 (2014.01); G06T 3/4046 (2024.01); G06T 9/00 (2006.01)
CPC H04N 19/30 (2014.11) [G06T 3/4046 (2013.01); G06T 9/002 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving video content to be displayed by a computing device;
based on a video content type associated with the video content and one or more neural networks pre-trained for the video content type to output parameters used to optimize downscaling and upscaling of video, downscaling the video content into one or more layers associated with one or more residual signals;
encoding a base layer of the one or more layers;
based on the encoded base layer, the one or more residual signals, and the one or more neural networks, upscaling the video content into the one or more layers of the video content;
encoding the one or more residual signals;
determining, based on the video content type and an analysis of one or more features extracted from the video content during the downscaling and the upscaling, one or more parameters associated with the one or more layers; and
sending, to the computing device, the one or more parameters, the encoded one or more residual signals, and the encoded base layer.