| CPC H04L 1/248 (2013.01) [G06N 3/042 (2023.01); G06N 3/045 (2023.01); G06T 7/0002 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/20048 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30168 (2013.01)] | 36 Claims |

|
1. A method for assessing transmission errors in a visual media input, comprising:
obtaining domain knowledge from the visual media input by content analysis, codec analysis, distortion analysis, and/or human visual system (HVS) modeling;
dividing the visual media input into partitions;
passing each partition into deep neural networks (DNNs) that produce DNN outputs indicating a transmission error assessment of the respective partition; and
combining the DNN outputs of the partitions with the domain knowledge to produce an overall assessment of the transmission errors in the visual media input.
|
|
19. A system for assessing transmission errors in a visual media input, comprising, a computing device programmed to:
obtain domain knowledge from the visual media input by content analysis, codec analysis, distortion analysis, and/or human visual system (HVS) modeling;
divide the visual media input into partitions;
pass each partition into deep neural networks (DNNs) that produce DNN outputs indicating a transmission error assessment of the respective partition; and
combine the DNN outputs of the partitions with the domain knowledge to produce an overall assessment of the transmission errors in the visual media input.
|