Abstract: In various examples, machine learning of encoding parameter values for a network is performed using a video encoder. Feedback associated with streaming video encoded by a video encoder over a network may be applied to an MLM(s). Using such feedback, the MLM(s) may predict a value(s) of an encoding parameter(s). The video encoder may then use the value to encode subsequent video data for the streaming. By using the video encoder in training, the MLM(s) may learn based on actual encoded parameter values of the video encoder. The MLM(s) may be trained via reinforcement learning based on video encoded by the video encoder. A rewards metric(s) may be used to train the MLM(s) using data generated or applied to the physical network in which the MLM(s) is to be deployed and/or a simulation thereof. Penalty metric(s) (e.g., the quantity of dropped frames) may also be used to train the MLM(s).

Abstract: A method includes determining, through a processor and/or a hardware engine, edge pixels and flat pixels of a video frame of a video sequence during decoding thereof or post-processing associated with the decoding based on a predetermined threshold, and quantifying spatial correlation of pixels of the video frame around edges thereof to estimate strength of ringing artifacts and spatial and temporal persistence thereof across the video frame and across video frames of the video sequence. The method also includes adaptively and spatially filtering the pixels around the edges of the video frame, adaptively and temporally filtering the video frame, and blending an output of the adaptive spatial filtering and the adaptive temporal filtering to generate an output with suppressed ringing artifacts, spatial and temporal persistence thereof and artifacts resulting from the cumulative effect of compression therein.

Abstract: A method includes determining, through a processor and/or a hardware engine, edge pixels and flat pixels of a video frame of a video sequence during decoding thereof or post-processing associated with the decoding based on a predetermined threshold, and quantifying spatial correlation of pixels of the video frame around edges thereof to estimate strength of ringing artifacts and spatial and temporal persistence thereof across the video frame and across video frames of the video sequence. The method also includes adaptively and spatially filtering the pixels around the edges of the video frame, adaptively and temporally filtering the video frame, and blending an output of the adaptive spatial filtering and the adaptive temporal filtering to generate an output with suppressed ringing artifacts, spatial and temporal persistence thereof and artifacts resulting from the cumulative effect of compression therein.