Self-Supervised Monocular Image Depth Learning and Confidence Estimation

TL;DR

This paper introduces a self-supervised framework for monocular image depth estimation that uses a patch-based cost function with ZNCC to improve accuracy and provides confidence estimation to enhance robustness.

Contribution

It presents a novel patch-based cost function using ZNCC for self-supervised depth learning and confidence estimation from monocular images.

Findings

Outperforms state-of-the-art on KITTI dataset

Improves depth estimation accuracy and robustness

Provides reliable confidence maps for depth predictions

Abstract

Convolutional Neural Networks (CNNs) need large amounts of data with ground truth annotation, which is a challenging problem that has limited the development and fast deployment of CNNs for many computer vision tasks. We propose a novel framework for depth estimation from monocular images with corresponding confidence in a self-supervised manner. A fully differential patch-based cost function is proposed by using the Zero-Mean Normalized Cross Correlation (ZNCC) that takes multi-scale patches as a matching strategy. This approach greatly increases the accuracy and robustness of the depth learning. In addition, the proposed patch-based cost function can provide a 0 to 1 confidence, which is then used to supervise the training of a parallel network for confidence map learning and estimation. Evaluation on KITTI dataset shows that our method outperforms the state-of-the-art results.