US 12,470,711 B2
Content-adaptive signal-to noise-ratio-guided two-pass video encoding for cloud gaming
Richard Xie, Cupertino, CA (US); Ramachandra Tahasildar, Cupertino, CA (US); Alex Sukhanov, Sunnyvale, CA (US); Danny Hong, New York, NY (US); and Beril Erkin, Mountain View, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Appl. No. 18/033,473
Filed by GOOGLE LLC, Mountain View, CA (US)
PCT Filed Nov. 25, 2020, PCT No. PCT/US2020/062161
§ 371(c)(1), (2) Date Apr. 24, 2023,
PCT Pub. No. WO2022/115102, PCT Pub. Date Jun. 2, 2022.
Prior Publication US 2023/0396766 A1, Dec. 7, 2023
Int. Cl. H04N 19/124 (2014.01); A63F 13/355 (2014.01); H04N 19/132 (2014.01); H04N 19/136 (2014.01); H04N 19/172 (2014.01); H04N 19/42 (2014.01); H04N 19/463 (2014.01)
CPC H04N 19/124 (2014.11) [A63F 13/355 (2014.09); H04N 19/132 (2014.11); H04N 19/136 (2014.11); H04N 19/172 (2014.11); H04N 19/42 (2014.11); H04N 19/463 (2014.11)] 20 Claims
OG exemplary drawing
 
1. A method for encoding video, comprising:
for each frame in a stream of frames:
performing a first encoding pass of a downscaled version of the frame;
determining an estimated peak-signal-to-noise ratio for the frame based on the first encoding pass of the downscaled version of the frame;
determining, by a statistical model, a quantization parameter in response to the estimated peak-signal-to-noise ratio of the frame being greater than a target peak-signal-to-noise ratio for the stream of frames; and
encoding the frame in a second encoding pass based on the quantization parameter.