US 12,126,828 B2
Video encoding through non-saliency compression for live streaming of high definition videos in low-bandwidth transmission
Umar Asif, Melbourne (AU); Lenin Mehedy, Doncaster East (AU); and Jianbin Tang, Doncaster East (AU)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Jul. 30, 2023, as Appl. No. 18/361,887.
Application 18/361,887 is a continuation of application No. 17/105,356, filed on Nov. 25, 2020, granted, now 11,758,182.
Prior Publication US 2024/0022759 A1, Jan. 18, 2024
Int. Cl. H04N 19/54 (2014.01); G06N 20/00 (2019.01); G06V 10/46 (2022.01); H04N 19/23 (2014.01); H04N 21/234 (2011.01)
CPC H04N 19/54 (2014.11) [G06N 20/00 (2019.01); G06V 10/462 (2022.01); H04N 19/23 (2014.11); H04N 21/23418 (2013.01)] 9 Claims
OG exemplary drawing
 
1. A computer-implemented method of decoding video data in multiple resolution formats, the computer-implemented method comprising:
receiving an encoded video stream including a salient data and a non-salient data, the salient data having a higher resolution format than the non-salient data;
receiving one or more of a link to access operation of, or load executable code for, a Generative Adversarial Network (GAN) machine learning model trained to identify non-salient features based on previously recorded video streams;
decoding the video stream into the non-salient data in a lower-resolution format and the salient data in the higher resolution format;
reconstructing the non-salient data to a higher resolution format using the GAN machine learning model; and
combining the salient data and the reconstructed non-salient data to form a video stream in the higher-resolution format of the salient data.