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).

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).

A picture encoding/decoding method and a related apparatus are provided. The picture decoding method includes obtaining a current picture; selecting, from a knowledge base, K reference pictures of the current picture, where at least one picture in the knowledge base does not belong to a random access segment in which the current picture is located and wherein K is an integer greater than or equal to 1; and decoding the current picture according to the K reference pictures.

One embodiment of the present invention sets forth a technique for detecting a shot change in a video. The technique includes calculating a first plurality of difference values for pixel blocks included in a first video frame and a second video frame and calculating a first cumulative distribution function (CDF) based on the first plurality of difference values. The technique further includes calculating a second plurality of difference values for pixel blocks included in a first plurality of video frames that is contiguous with the first video frame and calculating a second CDF based on the second plurality of difference values. The technique further includes comparing the first CDF to the second CDF to detect a shot change.

An image processing method is provided. The method includes acquiring a video. The method includes using an object detection engine to detect a person in the video. The object detection engine is integrated with an image signal processing pipeline. The method includes transmitting the video over a network. The method includes determining that the detected person has moved less than a pre-set distance. The method includes, responsive to the determining, pausing transmission of the video. An embedded image processor including an object detection engine is also provided.

An image processing method is provided. The method includes acquiring a video. The method includes using an object detection engine to detect a person in the video. The object detection engine is integrated with an image signal processing pipeline. The method includes transmitting the video over a network. The method includes determining that the detected person has moved less than a pre-set distance. The method includes, responsive to the determining, pausing transmission of the video. An embedded image processor including an object detection engine is also provided.

The present disclosure provides a method and an apparatus for coding a video, a device and a medium. An estimated quantization parameter of a target picture in a target GOP is determined. An inter-picture change degree of the target GOP is determined. A propagation ratio of a target picture block to a reference picture block in the target picture is determined based on the estimated quantization parameter of the target picture and the inter-picture change degree. The video is coded based on the propagation ratio.

A video encoding apparatus includes a content activity analyzer circuit and a video encoder circuit. The content activity analyzer circuit applies a content activity analysis process to consecutive frames, to generate content activity analysis results. The consecutive frames are derived from input frames of the video encoding apparatus. The content activity analysis process includes: deriving a first content activity analysis result according to a first frame and a second frame in the consecutive frames, wherein the first content activity analysis result includes a processed frame distinct from the second frame; and deriving a second content activity analysis result according to a third frame included in the consecutive frames and the processed frame. The video encoder circuit performs a video encoding process to generate a bitstream output of the video encoding apparatus, wherein information derived from the content activity analysis results is referenced by the video encoding process.

A method for encoding a multi-view frame in a video encoder is provided that includes computing a depth quality sensitivity measure for a multi-view coding block in the multi-view frame, computing a depth-based perceptual quantization scale for a 2D coding block of the multi-view coding block, wherein the depth-based perceptual quantization scale is based on the depth quality sensitive measure and a base quantization scale for the 2D frame including the 2D coding block, and encoding the 2D coding block using the depth-based perceptual quantization scale.

A method for encoding a multi-view frame in a video encoder is provided that includes computing a depth quality sensitivity measure for a multi-view coding block in the multi-view frame, computing a depth-based perceptual quantization scale for a 2D coding block of the multi-view coding block, wherein the depth-based perceptual quantization scale is based on the depth quality sensitive measure and a base quantization scale for the 2D frame including the 2D coding block, and encoding the 2D coding block using the depth-based perceptual quantization scale.