RRL:Regional Rotation Layer in Convolutional Neural Networks

TL;DR

This paper introduces RRL, a parameter-free module that enhances rotation invariance in CNNs without increasing complexity, improving performance on rotated images especially in fields with limited upright data.

Contribution

The proposed RRL module can be integrated into existing CNNs to achieve rotation invariance without additional parameters or training complexity.

Findings

Effective on LeNet-5, ResNet-18, tiny-yolov3

Performs well on rotated test sets

Suitable for biomedical and astronomical applications

Abstract

Convolutional Neural Networks (CNNs) perform very well in image classification and object detection in recent years, but even the most advanced models have limited rotation invariance. Known solutions include the enhancement of training data and the increase of rotation invariance by globally merging the rotation equivariant features. These methods either increase the workload of training or increase the number of model parameters. To address this problem, this paper proposes a module that can be inserted into the existing networks, and directly incorporates the rotation invariance into the feature extraction layers of the CNNs. This module does not have learnable parameters and will not increase the complexity of the model. At the same time, only by training the upright data, it can perform well on the rotated testing set. These advantages will be suitable for fields such as biomedicine and astronomy where it is difficult to obtain upright samples or the target has no directionality. Evaluate our module with LeNet-5, ResNet-18 and tiny-yolov3, we get impressive results.