Visual quality metrics
In one embodiment, a method includes receiving multimedia content information associated with at least one segment of a multimedia content, receiving a request to view the at least one segment of the multimedia content from a client device, logging playback information associated with the viewing of the at least one segment of the multimedia content, determining a multimedia quality metric associated with the at least one segment of multimedia content based in part upon a portion of the received multimedia content information and a portion of the logged playback information, and classifying the at least one segment of the multimedia content with the multimedia quality metric.
1. A method, comprising:
by a computer server machine, receiving multimedia content information associated with at least one segment of a multimedia content;
by the computer server machine, receiving, from a client device, a request to view the at least one segment of the multimedia content;
by the computer server machine, logging playback information associated with the viewing of the at least one segment of the multimedia content;
by the computer server machine, determining a multimedia quality metric associated with the at least one segment of multimedia content using a machine-learning model trained to compute the multimedia quality metric based in part upon a portion of the received multimedia content information and a portion of the logged playback information; and
by the computer server machine, classifying the at least one segment of the multimedia content with the multimedia quality metric.
2. The method of claim 1 , further comprising determining, by the computer server machine, that the at least one segment of the multimedia content is a high-quality segment by comparing the multimedia quality metric with a threshold.
3. The method of claim 2 , wherein the threshold comprises an adaptive threshold based in part upon a geographical region associated with the client device.
4. The method of claim 1 , further comprising:
receiving feedback associated with the classification of the at least one segment of the multimedia content; and
updating a calculation used to determine the multimedia quality metric.
5. The method of claim 1 , wherein the multimedia content information may comprise at least one of the following characteristics:
resolution of the at least one segment of the multimedia content;
content analytics associated with the at least one segment of the multimedia content;
compression of the at least one segment of the multimedia content;
encoding format of the at least one segment of the multimedia content; and
compression quality associated with the at least one segment of the multimedia content.
6. The method of claim 1 , wherein the playback information may comprise at least one of the following characteristics:
bit rate associated with the viewing of the at least one segment of the multimedia content;
dimension of the display associated with the client device;
orientation of the display associated with the viewing of the at least one segment of the multimedia content;
resolution of display associated with the client device; and
viewing interface associated with the viewing of the at least one segment of the multimedia content.
7. The method of claim 1 , wherein receiving multimedia content information associated with at least one segment of a multimedia content further comprises receiving multimedia content information associated with at least one segment of a multimedia content from one or more transcoding servers.
8. The method of claim 1 , wherein the machine-learning model is trained with a large collection of multimedia content streaming records.
9. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
receive multimedia content information associated with at least one segment of a multimedia content;
receive, from a client device, a request to view the at least one segment of the multimedia content;
log playback information associated with the viewing of the at least one segment of the multimedia content;
determine a multimedia quality metric associated with the at least one segment of multimedia content using a machine-learning model trained to compute the multimedia quality metric based in part upon a portion of the received multimedia content information and a portion of the logged playback information; and
classify the at least one segment of the multimedia content with the multimedia quality metric.
10. The media of claim 9 , wherein the software is further operable when executed to determine that the at least one segment of the multimedia content is a high-quality segment by comparing the multimedia quality metric with a threshold.
11. The media of claim 10 , wherein the threshold comprises an adaptive threshold based in part upon a geographical region associated with the client device.
12. The media of claim 9 , wherein the software is further operable when executed to:
receive feedback associated with the classification of the at least one segment of the multimedia content; and
update a calculation used to determine the multimedia quality metric.
13. The media of claim 9 , wherein the multimedia content information may comprise at least one of the following characteristics:
resolution of the at least one segment of the multimedia content;
content analytics associated with the at least one segment of the multimedia content;
compression of the at least one segment of the multimedia content;
encoding format of the at least one segment of the multimedia content; and
compression quality associated with the at least one segment of the multimedia content.
14. The media of claim 9 , wherein the playback information may comprise at least one of the following characteristics:
bit rate associated with the viewing of the at least one segment of the multimedia content;
dimension of the display associated with the client device;
orientation of the display associated with the viewing of the at least one segment of the multimedia content;
resolution of display associated with the client device; and
viewing interface associated with the viewing of the at least one segment of the multimedia content.
15. The media of claim 9 , wherein receiving multimedia content information associated with at least one segment of a multimedia content further comprises receiving multimedia content information associated with at least one segment of a multimedia content from one or more transcoding servers.
16. The media of claim 9 , wherein the machine-learning model is trained with a large collection of multimedia content streaming records.
17. A system comprising:
one or more processors; and
one or more computer-readable non-transitory storage media coupled to one or more of the processors and comprising instructions operable when executed by one or more of the processors to cause the system to:
receive multimedia content information associated with at least one segment of a multimedia content;
receive, from a client device, a request to view the at least one segment of the multimedia content;
log playback information associated with the viewing of the at least one segment of the multimedia content;
determine a multimedia quality metric associated with the at least one segment of multimedia content using a machine-learning model trained to compute the multimedia quality metric based in part upon a portion of the received multimedia content information and a portion of the logged playback information; and
classify the at least one segment of the multimedia content with the multimedia quality metric.
18. The system of claim 17 , wherein the processors are further operable when executing the instructions to determine that the at least one segment of the multimedia content is a high-quality segment by comparing the multimedia quality metric with a threshold.