Patent attributes
Provided a video processing method, a device and electronic equipment, which can process a video including multiple human body objects to obtain a plot video segment for any one of the multiple human body objects. The embodiments of the present application can carry out pertinent observations on the human body objects and improve the sensory experience of video viewers. The video processing method includes obtaining a first video including multiple human body objects; determining a detection region of a first object among the multiple human body objects according to at least one image frame of the first video; performing human behavior feature detection on the detection region of the first object in the first video to obtain a first plot video segment of the first object in the first video.
