Semi-Supervised Deep Learning for Monocular Depth Map Prediction

TL;DR

This paper introduces a semi-supervised deep learning method for monocular depth map prediction that leverages sparse ground truth and stereo image consistency, achieving superior results over existing methods.

Contribution

It presents a novel semi-supervised approach combining sparse ground truth with stereo consistency for improved monocular depth estimation.

Findings

Outperforms state-of-the-art depth prediction methods

Effective use of sparse LiDAR data and stereo consistency

Demonstrates robustness in outdoor dynamic environments

Abstract

Supervised deep learning often suffers from the lack of sufficient training data. Specifically in the context of monocular depth map prediction, it is barely possible to determine dense ground truth depth images in realistic dynamic outdoor environments. When using LiDAR sensors, for instance, noise is present in the distance measurements, the calibration between sensors cannot be perfect, and the measurements are typically much sparser than the camera images. In this paper, we propose a novel approach to depth map prediction from monocular images that learns in a semi-supervised way. While we use sparse ground-truth depth for supervised learning, we also enforce our deep network to produce photoconsistent dense depth maps in a stereo setup using a direct image alignment loss. In experiments we demonstrate superior performance in depth map prediction from single images compared to the state-of-the-art methods.