1 paper
Nan Yang, Laicheng Zhong, Fan Huang +2
The convolutional neural network (CNN) learns the same object in different positions in images, which can improve the recognition accuracy of the model. An implication of this is t…