End-to-End Learning of Geometry and Context for Deep Stereo Regression

TL;DR

This paper introduces a deep learning approach for stereo disparity estimation that uses geometric knowledge, a cost volume, and 3D convolutions to achieve high accuracy and speed without post-processing.

Contribution

It presents a novel end-to-end deep learning architecture utilizing a differentiable soft argmin for disparity regression, improving accuracy and efficiency over previous methods.

Findings

Achieved state-of-the-art results on KITTI dataset.

Significantly faster inference compared to existing methods.

End-to-end training without post-processing or regularization.

Abstract

We propose a novel deep learning architecture for regressing disparity from a rectified pair of stereo images. We leverage knowledge of the problem's geometry to form a cost volume using deep feature representations. We learn to incorporate contextual information using 3-D convolutions over this volume. Disparity values are regressed from the cost volume using a proposed differentiable soft argmin operation, which allows us to train our method end-to-end to sub-pixel accuracy without any additional post-processing or regularization. We evaluate our method on the Scene Flow and KITTI datasets and on KITTI we set a new state-of-the-art benchmark, while being significantly faster than competing approaches.