High-performance Semantic Segmentation Using Very Deep Fully Convolutional Networks

TL;DR

This paper introduces a high-performance semantic segmentation method using very deep residual networks, incorporating novel training techniques and network configurations to achieve state-of-the-art accuracy on benchmark datasets.

Contribution

The paper presents a new approach to semantic segmentation with very deep residual networks, including a simulation method for high-resolution features and an online bootstrapping training strategy.

Findings

Achieved 78.3% mean IoU on PASCAL VOC 2012

Outperformed previous methods on Cityscapes dataset

Demonstrated effectiveness of online bootstrapping and residual dropout

Abstract

We propose a method for high-performance semantic image segmentation (or semantic pixel labelling) based on very deep residual networks, which achieves the state-of-the-art performance. A few design factors are carefully considered to this end. We make the following contributions. (i) First, we evaluate different variations of a fully convolutional residual network so as to find the best configuration, including the number of layers, the resolution of feature maps, and the size of field-of-view. Our experiments show that further enlarging the field-of-view and increasing the resolution of feature maps are typically beneficial, which however inevitably leads to a higher demand for GPU memories. To walk around the limitation, we propose a new method to simulate a high resolution network with a low resolution network, which can be applied during training and/or testing. (ii) Second, we propose an online bootstrapping method for training. We demonstrate that online bootstrapping is critically important for achieving good accuracy. (iii) Third we apply the traditional dropout to some of the residual blocks, which further improves the performance. (iv) Finally, our method achieves the currently best mean intersection-over-union 78.3\% on the PASCAL VOC 2012 dataset, as well as on the recent dataset Cityscapes.