Improving Prediction Accuracy of Semantic Segmentation Methods Using Convolutional Autoencoder Based Pre-processing Layers

TL;DR

This paper introduces a convolutional autoencoder-based pre-processing layer to enhance the prediction accuracy of semantic segmentation methods like FCN, demonstrating significant improvements in generalization and accuracy with minimal parameter increase.

Contribution

The paper presents a novel, simple pre-processing approach using autoencoder layers that significantly improves semantic segmentation accuracy and generalization, applicable to various methods.

Findings

Mean IoU increased by 18.7% with the proposed method

Significant improvement in generalization ability observed

Method is simple, with minimal parameter increase

Abstract

In this paper, we propose a method to improve prediction accuracy of semantic segmentation methods as follows: (1) construct a neural network that has pre-processing layers based on a convolutional autoencoder ahead of a semantic segmentation network, and (2) train the entire network initialized by the weights of the pre-trained autoencoder. We applied this method to the fully convolutional network (FCN) and experimentally compared its prediction accuracy on the cityscapes dataset. The Mean IoU of the proposed target model with the He normal initialization is 18.7% higher than that of FCN with the He normal initialization. In addition, those of the modified models of the target model are significantly higher than that of FCN with the He normal initialization. The accuracy and loss curves during the training showed that these are resulting from the improvement of the generalization ability. All of these results provide strong evidence that the proposed method is significantly effective in improving the prediction accuracy of FCN. The proposed method has the following features: it is comparatively simple, whereas the effect on improving the generalization ability and prediction accuracy of FCN is significant; the increase in the number of parameters by using it is very small, and that in the computation time is substantially large. In principle, the proposed method can be applied to other semantic segmentation methods. For semantic segmentation, at present, there is no effective way to improve the prediction accuracy of existing methods. None have published a method which is the same as or similar to our method and none have used such a method in practice. Therefore, we believe that our method is useful in practice and worthy of being widely known and used.