Symmetry-aware Depth Estimation using Deep Neural Networks

TL;DR

This paper introduces a symmetry-aware neural network architecture that leverages symmetry information to significantly improve single-view depth estimation accuracy, especially for man-made objects.

Contribution

The paper presents a novel CNN architecture that estimates symmetric correspondences and uses this information to enhance depth prediction quality.

Findings

Outperforms state-of-the-art depth estimation methods

Utilizes symmetry to improve depth accuracy for man-made objects

Demonstrates significant quality improvements through extensive experiments

Abstract

Due to the abundance of 2D product images from the Internet, developing efficient and scalable algorithms to recover the missing depth information is central to many applications. Recent works have addressed the single-view depth estimation problem by utilizing convolutional neural networks. In this paper, we show that exploring symmetry information, which is ubiquitous in man made objects, can significantly boost the quality of such depth predictions. Specifically, we propose a new convolutional neural network architecture to first estimate dense symmetric correspondences in a product image and then propose an optimization which utilizes this information explicitly to significantly improve the quality of single-view depth estimations. We have evaluated our approach extensively, and experimental results show that this approach outperforms state-of-the-art depth estimation techniques.