Wider or Deeper: Revisiting the ResNet Model for Visual Recognition
作者：
Highlights：
• 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.
摘要
•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.
论文关键词：Image classification,Semantic segmentation,Residual network
论文评审过程：Received 24 June 2018, Revised 4 November 2018, Accepted 4 January 2019, Available online 6 January 2019, Version of Record 28 January 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.01.006
