Efficient and Invariant Convolutional Neural Networks for Dense Prediction

TL;DR

This paper introduces methods to make convolutional neural networks invariant to rotation and flip transformations, specifically for dense prediction tasks like image segmentation, by using kernel rotation and flip techniques.

Contribution

The paper proposes novel kernel rotation and flip methods to achieve rotation and flip invariance in CNNs for dense prediction tasks, addressing limitations of previous invariance approaches.

Findings

Achieves rotation and flip invariance with minimal resource increase

Uses maxout to combine multiple feature maps efficiently

Experimental results show effective invariance with reasonable memory and time costs

Abstract

Convolutional neural networks have shown great success on feature extraction from raw input data such as images. Although convolutional neural networks are invariant to translations on the inputs, they are not invariant to other transformations, including rotation and flip. Recent attempts have been made to incorporate more invariance in image recognition applications, but they are not applicable to dense prediction tasks, such as image segmentation. In this paper, we propose a set of methods based on kernel rotation and flip to enable rotation and flip invariance in convolutional neural networks. The kernel rotation can be achieved on kernels of 3 $\times$ 3, while kernel flip can be applied on kernels of any size. By rotating in eight or four angles, the convolutional layers could produce the corresponding number of feature maps based on eight or four different kernels. By using flip, the convolution layer can produce three feature maps. By combining produced feature maps using maxout, the resource requirement could be significantly reduced while still retain the invariance properties. Experimental results demonstrate that the proposed methods can achieve various invariance at reasonable resource requirements in terms of both memory and time.