US 12,284,364 B2
Machine learning (ML)-based rate control algorithm to find quantization parameter (QP)
Kannan Vairavan, San Mateo, CA (US); and Milan Mehta, San Mateo, CA (US)
Assigned to Sony Interactive Entertainment Inc.
Filed by Sony Interactive Entertainment Inc., Tokyo (JP)
Filed on Mar. 6, 2023, as Appl. No. 18/179,256.
Prior Publication US 2024/0305789 A1, Sep. 12, 2024
Int. Cl. H04N 19/149 (2014.01); H04N 19/172 (2014.01); H04N 19/184 (2014.01)
CPC H04N 19/149 (2014.11) [H04N 19/172 (2014.11); H04N 19/184 (2014.11)] 19 Claims
OG exemplary drawing
 
1. An apparatus comprising:
at least one processor system configured to:
input to at least a first machine learning (ML) model at least one non-intraframe (I-frame), at least one I-frame, and at least a representation of bits for a video frame;
provide output from the first ML model to at least a first video encoder of a first encoder type, the output comprising at least a quantization parameter (QP);
execute the first video encoder to output at least one video frame based at least in part on the QP; and
execute at least a second ML model to generate QPs for non-reference frames for a second encoder type such that one ML model per encoder type is provided.