Adaptive noise reduction engine for streaming video
AssigneeMicrosoft Technology Licensing, LLC
InventorsStacey Spears, Haoyun Wu, Wenbo Zhang, Sudhakar Prabhu
Life of this patent
Estimating the term…
Abstract
An adaptive video noise reduction (AVNR) engine reduces noise in streaming video. The engine reads embedded information in the streaming video describing attributes of the video, such as bit rate, frame rate, resolution, content type and quality of the streaming video. The information may be included in a manifest of the streaming video. Based on the video attributes, the engine outputs one or more input values to applications that reduce noise of the streaming video. The one or more applications may reduce mosquito, block, temporal, random and banding noise of the video in response to the input values from the engine.
Claims17
A method comprising:
- receiving data comprising video data representing a video and attribute data comprising a quality indicator value inserted into the video data during encoding of the video;
- determining one or more inputs to a first application for reducing noise in the video, the one or more inputs derived from at least the quality indicator value;
- selecting at least a first input of the one or more inputs to be input into the first application to reduce noise in the video;
- sending the at least first input to the first application, the first application operative to reduce noise in the video based on the at least first input;
- and
- sending at least a second input to a second application to enhance detail in the video, wherein the second input is based on at least the first input of the one or more inputs derived from at least the quality indicator value, the second application other than an application for reducing noise.
A computing device, comprising;
- one or more processors;
- and
- one or more processor readable storage memories holding instructions executable by the one or more processors to:
- receive data comprising video data representing a video and attribute data comprising a quality indicator value inserted into the video data during encoding of the video;
- determine one or more inputs to a first application for reducing noise in the video, the one or more inputs derived from at least the quality indicator value;
- select at least a first input of the one or more inputs to be input into the first application to reduce noise in the video;
- send the at least first input to the first application, the first application operative to reduce noise in the video based on the at least first input;
- and
- send at least a second input to a second application to enhance detail in the video, wherein the second input is based on at least the first input of the one or more inputs derived from at least the quality indicator value, the second application other than an application for reducing noise.
A method comprising:
- receiving data comprising video data representing a video;
- receiving attribute data representing at least one attribute of the video, the attribute data comprising a quality indicator value inserted into the video data during encoding of the video and at least one of a frame rate and a bit rate of the video;
- determining one or more inputs to a first application for reducing noise in the video, the one or more inputs derived from the quality indicator value and at least one of the frame rate and the bit rate of the video;
- selecting at least a first input of the one or more inputs to be input into the first application to reduce noise in the video;
- sending the at least first input to the first application, the first application operative to reduce noise in the video based on the at least first input;
- and
- sending at least a second input to a second application to enhance detail in the video, wherein the second input is based on at least the first input of the one or more inputs derived from at least the quality indicator value, the second application other than an application for reducing noise.
Dependencies are inferred from the claim text.