US 12,452,436 B2
Scene classification and learning for video compression
Weidong Mao, West Windsor, NJ (US); and Alexander Giladi, Princeton, NJ (US)
Assigned to Comcast Cable Communications, LLC, Philadelphia, PA (US)
Filed by Comcast Cable Communications, LLC, Philadelphia, PA (US)
Filed on Apr. 10, 2024, as Appl. No. 18/631,862.
Application 18/631,862 is a continuation of application No. 17/586,453, filed on Jan. 27, 2022, granted, now 11,985,337.
Application 17/586,453 is a continuation of application No. 16/291,076, filed on Mar. 4, 2019, granted, now 11,272,192, issued on Mar. 8, 2022.
Prior Publication US 2025/0080758 A1, Mar. 6, 2025
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 19/176 (2014.01); G06V 20/40 (2022.01); H04N 19/115 (2014.01); H04N 19/117 (2014.01); H04N 19/154 (2014.01); H04N 19/162 (2014.01); H04N 19/166 (2014.01); H04N 19/17 (2014.01); H04N 19/177 (2014.01); H04N 19/179 (2014.01); H04N 19/192 (2014.01); H04N 19/196 (2014.01); H04N 19/46 (2014.01); H04N 19/517 (2014.01); H04N 19/61 (2014.01); H04N 21/647 (2011.01)
CPC H04N 19/176 (2014.11) [G06V 20/49 (2022.01); H04N 19/166 (2014.11); H04N 19/517 (2014.11); H04N 21/64738 (2013.01); H04N 19/115 (2014.11); H04N 19/117 (2014.11); H04N 19/154 (2014.11); H04N 19/162 (2014.11); H04N 19/17 (2014.11); H04N 19/177 (2014.11); H04N 19/179 (2014.11); H04N 19/192 (2014.11); H04N 19/196 (2014.11); H04N 19/46 (2014.11); H04N 19/61 (2014.11)] 17 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a computing device, a content item and information indicating a plurality of visual elements in the content item;
allocating, to a plurality of regions of the content item and based on the plurality of visual elements in the content item, different encoding parameters;
determining that a first visual element of the plurality of visual elements moves from a first region of the plurality of regions to a second region of the plurality of regions based on one or more of:
an indication that motion, associated with the first visual element, is expected to be predictable; or
a speed of motion associated with the first visual element, wherein the second region is occupied by a second visual element of the plurality of visual elements;
adjusting, based on an amount of the first visual element that is visible in the second region, encoder parameters corresponding to the second region; and
causing the second region to be encoded using the adjusted encoder parameters.