Wasserstein Distances for Stereo Disparity Estimation

TL;DR

This paper introduces a novel neural network architecture and a Wasserstein distance-based loss function for stereo disparity and depth estimation, significantly improving accuracy especially around object boundaries and advancing 3D object detection.

Contribution

The paper presents a new neural network and loss function that enable arbitrary depth output and reduce errors in ambiguous regions, outperforming existing methods.

Findings

Reduces disparity estimation errors around object boundaries

Achieves state-of-the-art results in 3D object detection

Improves accuracy in ambiguous regions

Abstract

Existing approaches to depth or disparity estimation output a distribution over a set of pre-defined discrete values. This leads to inaccurate results when the true depth or disparity does not match any of these values. The fact that this distribution is usually learned indirectly through a regression loss causes further problems in ambiguous regions around object boundaries. We address these issues using a new neural network architecture that is capable of outputting arbitrary depth values, and a new loss function that is derived from the Wasserstein distance between the true and the predicted distributions. We validate our approach on a variety of tasks, including stereo disparity and depth estimation, and the downstream 3D object detection. Our approach drastically reduces the error in ambiguous regions, especially around object boundaries that greatly affect the localization of objects in 3D, achieving the state-of-the-art in 3D object detection for autonomous driving. Our code will be available at https://github.com/Div99/W-Stereo-Disp.