RegNet: Self-Regulated Network for Image Classification
This work provides an incremental improvement to ResNet architectures for image classification tasks by enhancing feature exploration.
This paper introduces RegNet, a self-regulated network that enhances ResNet architectures by incorporating a regulator module to extract complementary features. This module, composed of convolutional RNNs, addresses the limitation of simple shortcut connections in re-exploring new features. RegNet demonstrates promising performance on three image classification datasets compared to standard ResNet, SE-ResNet, and other state-of-the-art architectures.
The ResNet and its variants have achieved remarkable successes in various computer vision tasks. Despite its success in making gradient flow through building blocks, the simple shortcut connection mechanism limits the ability of re-exploring new potentially complementary features due to the additive function. To address this issue, in this paper, we propose to introduce a regulator module as a memory mechanism to extract complementary features, which are further fed to the ResNet. In particular, the regulator module is composed of convolutional RNNs (e.g., Convolutional LSTMs or Convolutional GRUs), which are shown to be good at extracting Spatio-temporal information. We named the new regulated networks as RegNet. The regulator module can be easily implemented and appended to any ResNet architecture. We also apply the regulator module for improving the Squeeze-and-Excitation ResNet to show the generalization ability of our method. Experimental results on three image classification datasets have demonstrated the promising performance of the proposed architecture compared with the standard ResNet, SE-ResNet, and other state-of-the-art architectures.