Video quality monitoring
View Patent ↗Systems and methods of perceptual quality monitoring of video information, communications, and entertainment that can estimate the perceptual quality of video with high accuracy, and can be used to produce quality scores that better correlate with subjective quality scores of an end user. The systems and methods of perceptual quality monitoring of video can generate, from an encoded input video bitstream, estimates of one or more quality parameters relating to the video, such as the coding bit rate parameter, the video frame rate parameter, and the packet loss rate parameter, and provide these video quality parameter estimates to a predetermined video quality estimation model. Because the estimates of the video quality parameters are generated from the encoded input video bitstream as it is being received, the systems and methods are suitable for use as QoE monitoring tools.
1. A method of monitoring a perceptual quality of video in a perceptual quality monitoring system, the perceptual quality monitoring system including a data collection component, a parameter estimation component, and a video quality estimation component, the method comprising the steps of:
receiving, at the data collection component, an encoded input video bitstream containing predetermined information and at least one video frame;
generating, by the parameter estimation component from at least some of the predetermined information contained in the encoded input video bitstream, a plurality of video quality parameter estimates including at least a frame rate parameter estimate and a packet loss rate parameter estimate;
generating, by the parameter estimation component, a coding bit rate parameter estimate as a function of the frame rate parameter estimate and the packet loss rate parameter estimate, the perceptual quality monitoring system further including a video frame complexity estimation component;
generating, by the video frame complexity estimation component, an estimate of the complexity of the at least one video frame using at least some of the predetermined information;
generating, by the parameter estimation component, a normalized version of the coding bit rate parameter estimate using the estimate of the complexity of the at least one video frame;
providing, by the parameter estimation component, indications of the normalized version of the coding bit rate parameter estimate, the frame rate parameter estimate, and the packet loss rate parameter estimate to the video quality estimation component; and
generating, by the video quality estimation component, an estimate of the perceptual quality of the video using at least the normalized version of coding bit rate parameter estimate, the frame rate parameter estimate, and the packet loss rate parameter estimate.
2. The method of claim 1 wherein the generating of the plurality of video quality parameter estimates includes extracting, by the parameter estimation component, the predetermined information from the encoded input video bitstream.
3. The method of claim 2 wherein the encoded input video bitstream further includes a plurality of video frames, and wherein the predetermined information includes at least one or more of a number of bits received, a number of packets received, and a number of packets lost, for a predetermined number of the plurality of video frames.
4. The method of claim 3 wherein the generating of the plurality of video quality parameter estimates includes calculating, by the parameter estimation component, at least one or more of the number of bits received, the number of packets received, and the number of packets lost, for the predetermined number of the plurality of video frames using a sliding window.
5. The method of claim 1 wherein the generating of the plurality of video quality parameter estimates includes calculating, by the parameter estimation component, one or more of:
a number of packets received in a predetermined number of video frames;
a number of packets received per video frame;
a number of bits received in the predetermined number of video frames;
a number of packets lost in the predetermined number of video frames;
a time interval between two adjacent video frames; and
a reference clock frequency for a predetermined coding format.
6. The method of claim 5 wherein the predetermined coding format is one of an H.264 coding format, an MPEG-2 coding format, an MPEG-4 coding format, a VC-1 coding format, and a VP8 coding format.
7. The method of claim 5 wherein the parameter estimation component includes a packet loss rate estimation component, and wherein the generating of the plurality of video quality parameter estimates includes generating, by the packet loss rate estimation component, the packet loss rate parameter estimate using the number of packets received and the number of packets lost.
8. The method of claim 5 wherein the parameter estimation component includes a frame rate estimation component, and wherein the generating of the plurality of video quality parameter estimates includes generating, by the frame rate estimation component, the frame rate parameter estimate using the time interval and the reference clock frequency.
9. The method of claim 5 wherein the parameter estimation component includes a coding bit rate estimation component, and wherein the generating of the coding bit rate parameter estimate includes generating, by the coding bit rate estimation component, the coding bit rate parameter estimate using the number of bits received and the number of packets received.
10. The method of claim 1 wherein the generating of the estimate of the complexity of the at least one video frame includes estimating, by the video frame complexity estimation component, a number of bits for use at a reference quantization level.
11. The method of claim 1 wherein the generating of the estimate of the complexity of the at least one video frame includes generating, by the video frame complexity estimation component, the estimate of the complexity of the at least one video frame using a plurality of actual bits for encoding the at least one video frame, and a quantization complexity factor.
12. A method of monitoring a perceptual quality video in a perceptual quality monitoring system, the perceptual quality monitoring system including a data collection component, a parameter estimation component, and a video quality estimation component, the method comprising the steps of:
receiving, at the data collection component, an encoded input video bitstream containing predetermined information and at least one video frame;
generating, by the parameter estimation component from at least some of the predetermined information contained in the encoded input video bitstream, a plurality of video quality parameter estimates including at least a frame rate parameter estimate and a packet loss rate parameter estimate;
generating, by the parameter estimation component, a coding bit rate parameter estimate as a function of the frame rate parameter estimate and the packet loss rate parameter estimate, the perceptual quality monitoring system further including a video frame complexity estimation component;
generating, by the video frame complexity estimation component, an estimate of the complexity of the at least one video frame using at least some of the predetermined information, the generating of the estimate of the complexity of the at least one video frame including:
generating the estimate of the complexity of the at least one video frame using a plurality of actual bits for encoding the at least one video frame, and a quantization complexity factor; and
generating the estimate of the complexity of the at least one video frame based on a number of coded macroblocks, a number of macroblocks in the respective video frame, and the quantization complexity factor; and
generating, by the video quality estimation component, an estimate of the perceptual quality of the video using at least the coding bit rate parameter estimate, the frame rate parameter estimate, and the packet loss rate parameter estimate.
13. A method of monitoring a perceptual quality of video in a perceptual quality monitoring system, the perceptual quality monitoring system including a data collection component, a parameter estimation component, and a video quality estimation component, the method comprising the steps of:
receiving, at the data collection component, an encoded input video bitstream containing predetermined information and at least one video frame;
generating, by the parameter estimation component from at least some of the predetermined information contained in the encoded input video bitstream, a plurality of video quality parameter estimates including at least a frame rate parameter estimate and a packet loss rate parameter estimate;
generating, by the parameter estimation component, a coding bit rate parameter estimate as a function of the frame rate parameter estimate and the packet loss rate parameter estimate, the perceptual quality monitoring system further including a video frame complexity estimation component;
generating, by the video frame complexity estimation component, an estimate of the complexity of the at least one video frame using at least some of the predetermined information, the generating of the estimate of the complexity of the at least one video frame including:
generating the estimate of the complexity of the at least one video frame using a plurality of actual bits for encoding the at least one video frame, and a quantization complexity factor; and
determinging the quantization complexity factor using an actual quantization step-size matrix and a reference quantization step-size matrix; and
generating, by the video quality estimation component, an estimate of the perceptual quality of the video using at least the coding bit rate parameter estimate, the frame rate parameter estimate, and the packet loss rate parameter estimate.
14. The method of claim 13 wherein the perceptual quality monitoring system further includes a coding bit rate normalization component, and wherein the method further includes generating, by the coding bit rate normalization component, a normalized version of the coding bit rate parameter estimate using the estimate of the complexity of the at least one video frame.
15. The method of claim 14 further comprising:
providing, by the parameter estimation component, indications of the packet loss rate parameter estimate, the frame rate parameter estimate, and the normalized version of the coding bit rate parameter estimate to the video quality estimation component.
16. The method of claim 15 wherein the generating of the estimate of the perceptual quality of the video includes generating, by the video quality estimation component, the estimate of the perceptual quality of the video using the packet loss rate parameter estimate, the frame rate parameter estimate, and the normalized version of the coding bit rate parameter estimate.
17. The method of claim 1 further including implementing, by the video quality estimation component, a predetermined video quality estimation model.
18. The method of claim 17 wherein the predetermined video quality estimation model is a G.1070 video quality estimation model.
19. A system for monitoring a perceptual quality of video, comprising:
a data collection component, a parameter estimation component, a video frame complexity estimation component, a coding bit rate normalization component, and a video quality estimation component,
wherein each of the data collection component, the parameter estimation component, the video frame complexity estimation component, the coding bit rate normalization component, and the video quality estimation component is embodied at least in part as computer readable code on a computer readable medium; and
at least one computerized device operative to execute the computer readable code from the computer readable medium:
to receive, at the data collection component, an encoded input video bitstream containing at least one video frame and predetermined information;
to generate, by the parameter estimation component from at least some of the predetermined information contained in the encoded input video bitstream, a plurality of video quality parameter estimates including at least a packet loss rate parameter estimate and a frame rate parameter estimate;
to generate, by the parameter estimation component, a coding bit rate parameter estimate as a function of the frame rate parameter estimate and the packet loss rate parameter estimate;
to generate, by the video frame complexity estimation component, an estimate of the complexity of the at least one video frame using at least some of the predetermined information;
to generate, by the a coding bit rate normalization component, a normalized version of the coding bit rate parameter estimate using the estimate of the complexity of the at least one video frame; and
to generate, by the a video quality estimation component, an estimate of the perceptual quality of the video using at least the packet loss rate parameter estimate, the frame rate parameter estimate, and the normalized version of the coding bit rate parameter estimate.
20. The system of claim 19 wherein the parameter estimation component is further operative to generate the coding bit rate parameter estimate using the estimate of the complexity of the at least one video frame.
21. The system of claim 19 wherein the video frame complexity estimation component is further operative to estimate a number of bits for use at a reference quantization level.
22. The system of claim 19 wherein the video frame complexity estimation component is further operative to generate the estimate of the complexity of the at least one video frame using a plurality of actual bits for encoding the at least one video frame, and a quantization complexity factor.
23. The system of claim 22 wherein the video frame complexity estimation component is further operative to generate the estimate of the complexity of the at least one video frame based on a number of coded macroblocks, a number of macroblocks in the respective video frame, and the quantization complexity factor.
24. The system of claim 22 wherein the video frame complexity estimation component is further operative to determine the quantization complexity factor using an actual quantization step-size matrix and a reference quantization step-size matrix.