Abstract: The present disclosure relates to systems and methods for video analysis. The systems may obtain a video stream. For each of a plurality of frames in the video stream, the systems may determine a reference score corresponding to the frame based on a spatiotemporal feature of the frame. The spatiotemporal feature of the frame may associate a feature of the frame with a feature of at least one adjacent frame of the frame. The systems may determine, from the plurality of frames in the video stream, one or more target frames based on reference scores corresponding to the plurality of frames.