Methods and apparatus to improve screen sharing based on identification of irrelevant video frames from interactive context

At least one non-transitory computer readable medium comprising instructions to cause at least one processor circuit to at least:

• execute a machine learning model based on a sequence of user input data and a corresponding sequence of application contextual data to determine an interaction classification associated with a screen sharing event initiated in a video conference, the interaction classification representing a content transition that is not intended to be rendered in the screen sharing event;
• identify, based on the interaction classification, video frames of the screen sharing event to exclude from a transport stream for the video conference;
• exclude the identified video frames from the transport stream;
• and
• provide remaining video frames of the screen sharing event to a video encoder, the video encoder to include the remaining video frames in the transport stream.

A method comprising:

• executing a machine learning model based on a sequence of user input data and a corresponding sequence of application contextual data to determine an interaction classification associated with a screen sharing event initiated in a video conference, the interaction classification represents a content transition that is not intended to be rendered in the screen sharing event;
• identifying, based on the interaction classification, video frames of the screen sharing event to exclude from a transport stream for the video conference;
• excluding the identified video frames from the transport stream;
• and
• including remaining video frames of the screen sharing event in the transport stream.

An apparatus comprising:

• machine learning model controller circuitry to process a sequence of input data and a corresponding sequence of application contextual data with a machine learning model to determine an interaction classification associated with a screen sharing event initiated in a video conference, the interaction classification represents a content transition that is not intended to be rendered in the screen sharing event;
• and
• frame selection circuitry to:
• identify, based on the interaction classification, video frames of the screen sharing event to exclude from a transport stream for the video conference;
• exclude the identified video frames from the transport stream;
• and
• provide remaining video frames of the screen sharing event to a video encoder, the video encoder to include the remaining video frames in the transport stream.