Reducing spatial redundancy in video recognition. SOTA computational efficiency.
blackfeather-wang
Back