Fast Disparity Estimation using Dense Networks

TL;DR

This paper introduces DenseMapNet, a compact and fast deep learning model for disparity estimation that effectively handles textureless and repetitive regions by incorporating dense network architecture and semantic reasoning.

Contribution

The paper presents DenseMapNet, a novel dense network-based CNN that reduces parameters and computational cost while maintaining high accuracy in disparity estimation.

Findings

DenseMapNet requires only 290k parameters.

It runs at 30Hz or faster on full-resolution stereo images.

Its accuracy is comparable to larger CNN-based methods.

Abstract

Disparity estimation is a difficult problem in stereo vision because the correspondence technique fails in images with textureless and repetitive regions. Recent body of work using deep convolutional neural networks (CNN) overcomes this problem with semantics. Most CNN implementations use an autoencoder method; stereo images are encoded, merged and finally decoded to predict the disparity map. In this paper, we present a CNN implementation inspired by dense networks to reduce the number of parameters. Furthermore, our approach takes into account semantic reasoning in disparity estimation. Our proposed network, called DenseMapNet, is compact, fast and can be trained end-to-end. DenseMapNet requires 290k parameters only and runs at 30Hz or faster on color stereo images in full resolution. Experimental results show that DenseMapNet accuracy is comparable with other significantly bigger CNN-based methods.