Reference-based video quality analysis-as-a-service (VQAaaS) for over-the-top (OTT) streaming
This disclosure provides for automated techniques to measure full reference-based QoE or VQA-as-a-Service (VQAaaS) for an Internet video stream. Generally, the approach herein involves pre-calculating VMAF scores for given media and then correlating those scores with VMAF scores computed from actual playback segments for the given media. By leveraging the pre-calculated VMAF scores and correlating them with playback data, the system provides for enhanced and accurate video quality analysis (VQA) to enable optimization of viewer Quality of Experience (QoE).
1 . A method of video quality measurement for a source video stream that is available for delivery in segments at multiple bitrates, comprising:
receiving a first data set, the first data set having been generated prior to streaming of the source video stream by pre-calculating a video quality metric per transcoded rendition, per segment, for the source video stream;
post-streaming of the source video stream:
receiving network log data, the network log data comprising a list of segments of the source video stream and in an order that the segments were selected and rendered by a media player, each segment in the list of segments including an identification of its transcoded rendition;
generating, as a second data set, the video quality metric for each of the segments in the list of segments;
comparing the video quality metrics in the first and second data sets; and
taking a given automated action in a streaming delivery platform based at least in part on a result of the comparison of the video quality metrics.
2 . The method as described in claim 1 wherein the video quality metric is Video Multimethod Assessment Fusion (VMAF).
3 . The method as described in claim 1 wherein the source video stream is an Over-The-Top (OTT) video stream and the delivery is one of: live streaming, and on-demand streaming.
4 . The method as described in claim 1 wherein the first data set is generated using a frame alignment algorithm that performs intra- and inter-frame alignment between each raw frame of the source video stream and the segments of each transcoded rendition.
5 . The method as described in claim 1 wherein the second data set is generated using a frame alignment algorithm that performs intra- and inter-frame alignment between the first frame of a rendition of the source video stream and an associated segment of a transcoded rendition.
6 . The method as described in claim 1 wherein the transcoded renditions are one of: HLS and DASH, and the delivery is Adaptive Bitrate (ABR) streaming.
7 . The method as described in claim 1 wherein the method is operated at least in part as-a-service.
8 . The method as described in claim 1 wherein the given action provisions additional resources or adjusts existing resources in the streaming delivery platform.
9 . The method as described in claim 1 wherein the first and second data sets are generated in part using an adjustment algorithm that measures Peak-Signal-To-Noise (PSTN) ratios of pixels within frames associated with the source video stream.
10 . The method as described in claim 1 , wherein generating the second data set further includes performing a segment number correction for at least segment in the list of segments.
11 . A Software-as-as-Service (SaaS) computing platform, comprising:
computing hardware; and
computer software executed on the computing hardware to provide video quality measurement for a source video stream that is available for delivery in segments at multiple bitrates, the computer software comprising program code configured to:
receive a first data set, the first data set having been generated prior to streaming of the source video stream by pre-calculating a video quality metric per transcoded rendition, per segment, for the source video stream;
post-streaming of the source video stream:
receive network log data, the network log data comprising a list of segments of the source video stream and in an order that the segments were selected and rendered by a media player, each segment in the list of segments including an identification of its transcoded rendition;
generate, as a second data set, the video quality metric for each of the segments in the list of segments;
compare the video quality metrics in the first and second data sets; and
take a given automated action in a streaming delivery platform based at least in part on a result of the comparison of the video quality metrics.
12 . The SaaS computing platform as described in claim 11 wherein the video quality metric is Video Multimethod Assessment Fusion (VMAF).
13 . The SaaS computing platform as described in claim 11 wherein the source video stream is an Over-The-Top (OTT) video stream and the delivery is one of: live streaming, and on-demand streaming.
14 . The SaaS computing platform as described in claim 11 , wherein the program code configured to generate the second data set further includes code that performs a segment number correction for at least one segment in the list of segments.
15 . A computer program product in a non-transitory computer readable medium, the computer program product holding computer program instructions that, when executed by one or more processors in a host processing system, provide a service, the computer program instructions comprising program code configured to:
receive a first data set, the first data set having been generated prior to streaming of the source video stream by pre-calculating a video quality metric per transcoded rendition, per segment, for the source video stream;
post-streaming of the source video stream;
receive network log data, the network log data comprising a list of segments of the source video stream and in an order that the segments were selected and rendered by a media player, each segment in the list of segments including an identification of its transcoded rendition;
generate, as a second data set, the video quality metric for each of the segments in the list of segments;
compare the video quality metrics in the first and second data sets; and
take a given automated action in a streaming delivery platform based at least in part on a result of the comparison of the video quality metrics.
16 . The computer program product as described in claim 15 , wherein the video quality metric is Video Multimethod Assessment Fusion (VMAF).
17 . The computer program product as described in claim 15 , wherein the source video stream is an Over-The-Top (OTT) video stream and the delivery is one of: live streaming, and on-demand streaming.
18 . The computer program product as described in claim 15 , wherein the program code configured to generate the second data set further includes code that performs a segment number correction for at least one segment in the list of segments.