US 11,870,833 B2
Methods and systems for encoder parameter setting optimization
Ching Yin Derek Pang, San Jose, CA (US); Kyrah Felder, Kennesaw, GA (US); Akshay Gadde, Fremont, CA (US); Paul Wilkins, Cambridge (GB); Cheng Chen, Milpitas, CA (US); and Yao-Chung Lin, Sunnyvale, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Aug. 31, 2021, as Appl. No. 17/462,591.
Prior Publication US 2023/0068026 A1, Mar. 2, 2023
Int. Cl. H04L 65/70 (2022.01); G06N 20/00 (2019.01); H04L 65/61 (2022.01); H04L 65/80 (2022.01); H04N 21/25 (2011.01)
CPC H04L 65/70 (2022.05) [G06N 20/00 (2019.01); H04L 65/61 (2022.05); H04L 65/80 (2013.01); H04N 21/251 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
identifying a media item to be provided to one or more users of a platform, wherein the media item is associated with a media class of a plurality of media classes, and wherein each of the plurality of media classes corresponds to at least one of one or more distinct image characteristics or a distinct content category;
obtaining encoding statistics associated with the identified media item based on an initial encoding of the identified media item by a media item encoder;
providing an indication of the identified media item and the obtained encoding statistics as input to a first machine learning model, wherein the first machine learning model is trained to predict, for a given media item, a set of encoder parameter settings that satisfy a performance criterion in view of a respective media class, of the plurality of media classes, associated with the given media item;
obtaining one or more outputs of the first machine learning model, wherein the one or more obtained outputs comprise encoder data identifying one or more sets of encoder parameter settings and, for each of the sets of encoder parameter settings, an indication of a level of confidence that a respective set of encoder parameter settings satisfies the performance criterion in view of the media class associated with the identified media item; and
causing the identified media item to be reencoded using the respective set of encoder parameter settings associated with the level of confidence that satisfies a confidence criterion.