Automated cinematic decisions based on descriptive models

One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

• at a first time period during an audio-video communication session:
• determine that a participant is located in an environment associated with the audio-video communication session;
• locate a reference body region of the participant;
• generate a reference color histogram of the reference body region, wherein the reference color histogram represents a distribution of one or more colors of the reference body region;
• at a second time period during the audio-video communication:
• determine that a subject participant is located in the environment;
• locate a body region of the subject participant from the second time period that corresponds with the reference body region of the participant from the first time period;
• generate a color histogram of the body region of the subject participant from the second time period, wherein the color histogram represents a distribution of one or more colors of the body region of the subject participant from the second time period;
• determine that the color histogram of the subject participant from the second time period matches the reference color histogram;
• based on the determination that the color histogram of the subject participant from the second time period matches the reference color histogram, identify the subject participant from the second time period as the same participant as the participant from the first time period.

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:
• at a first time period during an audio-video communication session:
• determine that a participant is located in an environment associated with the audio-video communication session;
• locate a reference body region of the participant;
• generate a reference color histogram of the reference body region, wherein the reference color histogram represents a distribution of one or more colors of the reference body region;
• at a second time period during the audio-video communication:
• determine that a subject participant is located in the environment;
• locate a body region of the subject participant from the second time period that corresponds with as the reference body region of the participant from the first time period;
• generate a color histogram of the body region of the subject participant from the second time period, wherein the color histogram represents a distribution of one or more colors of the body region of the subject participant from the second time period;
• determine that the color histogram of the subject participant from the second time period matches the reference color histogram;
• based on the determination that the color histogram of the subject participant from the second time period matches the reference color histogram, identify the subject participant from the second time period is the same participant as the participant from the first time period.