Objects removal from video framing

A non-transitory computer-readable medium comprising computer-executable instructions that, when executed by a processor, cause the processor to:

• detect a plurality of objects in captured video sensor data;
• identify a first object of the plurality of objects based on the captured video sensor data;
• identify a second object of the plurality of objects based on the captured video sensor data;
• determine a first liveness attribute of the first object based on the captured video sensor data;
• remove the first object from a teleconference framing based on the first liveness attribute;
• determine a second liveness attribute of the second object based on the captured video sensor data;
• and
• not remove the second object from the teleconference framing based on the second liveness attribute.

A computer-implemented method, the method comprising:

• identifying a first object that is within a field of view (FOV) of a video conferencing device using video data generated by the video conferencing device;
• identifying a second object that is within the FOV of the video conferencing device using the video data generated by the video conferencing device;
• determining a first liveness attribute associated with the first object using the video data generated by the video conferencing device, the first liveness attribute being indicative of whether the first object is a first living organic being;
• removing the first object from a video conference framing based on the first liveness attribute;
• determining a second liveness attribute associated with the second object using the video data generated by the video conferencing device, the second liveness attribute being indicative of whether the second object is a second living organic being;
• and
• not removing the second object from the video conference framing based on the second liveness attribute.