US 11,875,495 B2
Methods of video quality assessment using parametric and pixel level models
Xiaozhong Xu, State College, PA (US); Xiang Li, Saratoga, CA (US); and Shan Liu, San Jose, CA (US)
Assigned to TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed by TENCENT AMERICA LLC, Palo Alto, CA (US)
Filed on Jun. 29, 2021, as Appl. No. 17/362,145.
Claims priority of provisional application 63/063,712, filed on Aug. 10, 2020.
Prior Publication US 2022/0044385 A1, Feb. 10, 2022
Int. Cl. G06T 7/00 (2017.01); G06T 7/20 (2017.01)
CPC G06T 7/0002 (2013.01) [G06T 7/20 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/30168 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method of performing a video quality assessment (VQA), the method comprising:
receiving a bitstream comprising a plurality of images of a video, the plurality of images being divided into one or more groups;
identifying, from the bitstream, first images among the plurality of images to which a parametric-based VQA is to be applied;
determining bitstream level information based on applying the parametric-based VQA to each of the first images, the bitstream level information indicating any of a bitrate, a quantization parameter, and a frame rate;
identifying, from the bitstream, second images among the plurality of images to which a sample-based VQA is to be applied;
determining pixel level information based on applying the sample-based VQA to each of the second images;
assessing the video based on a combination of the bitstream level information and the pixel level information, the bitstream level information being selectively calibrated depending on the pixel level information; and
based on determining that the pixel level information is not available for at least one of the first images, calibrating the first score,
wherein the calibrating the first score comprises:
obtaining one or more first images prior to the at least one of the first images in a sequence in decoding order;
obtaining one or more second images corresponding to the one or more first images;
calculating a difference between a first scoring system of the parametric-based VQA and a second scoring system of the sample-based VQA based on the bitstream level information of the one or more first images and the pixel level information of the one or more second images; and
calibrating the bitstream level information based on the difference between the first scoring system and the second scoring system.