Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields

TL;DR

This paper introduces a deep convolutional neural field model that combines CNNs and continuous CRFs to improve depth estimation from single monocular images, outperforming existing methods.

Contribution

It presents a novel deep structured learning framework that jointly learns unary and pairwise potentials for depth estimation, including a faster fully convolutional network variant.

Findings

Outperforms state-of-the-art depth estimation methods on indoor and outdoor datasets.

The fully convolutional model with superpixel pooling is approximately 10 times faster.

Deeper networks lead to better depth estimation performance.

Abstract

In this article, we tackle the problem of depth estimation from single monocular images. Compared with depth estimation using multiple images such as stereo depth perception, depth from monocular images is much more challenging. Prior work typically focuses on exploiting geometric priors or additional sources of information, most using hand-crafted features. Recently, there is mounting evidence that features from deep convolutional neural networks (CNN) set new records for various vision applications. On the other hand, considering the continuous characteristic of the depth values, depth estimations can be naturally formulated as a continuous conditional random field (CRF) learning problem. Therefore, here we present a deep convolutional neural field model for estimating depths from single monocular images, aiming to jointly explore the capacity of deep CNN and continuous CRF. In particular, we propose a deep structured learning scheme which learns the unary and pairwise potentials of continuous CRF in a unified deep CNN framework. We then further propose an equally effective model based on fully convolutional networks and a novel superpixel pooling method, which is $\sim 10$ times faster, to speedup the patch-wise convolutions in the deep model. With this more efficient model, we are able to design deeper networks to pursue better performance. Experiments on both indoor and outdoor scene datasets demonstrate that the proposed method outperforms state-of-the-art depth estimation approaches.