1 paper
Yi Wei, Xinyu Pan, Hongwei Qin +2
In this paper, we propose a simple and general framework for training very tiny CNNs for object detection. Due to limited representation ability, it is challenging to train very ti…