Wider or Deeper: Revisiting the ResNet Model for Visual Recognition.

•We further develop the unravelled view of ResNets, which helps us better understand their behaviours. We demonstrate this in the context of a training process, which is the key difference from the original version 1.•We propose a group of relatively shallow convolutional networks based on our new understanding. Some of them perform comparably with the state-of-the-art approaches on the ImageNet classification dataset 2.•We evaluate the impact of using different networks on the performance of semantic image segmentation, and show these networks, as pre-trained features, can boost existing algorithms a lot.