DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation

TL;DR

DCPI-Depth introduces explicit geometric constraints and novel loss functions to improve unsupervised monocular depth estimation, especially in challenging scenarios like texture-less and dynamic regions, achieving state-of-the-art results.

Contribution

The paper presents three novel components—depth consistency loss, a differential property correlation loss, and a bidirectional co-adjustment strategy—that explicitly incorporate dense correspondence priors into depth learning.

Findings

Achieves state-of-the-art depth estimation accuracy.

Performs well in texture-less and dynamic regions.

Demonstrates improved smoothness and generalization.

Abstract

There has been a recent surge of interest in learning to perceive depth from monocular videos in an unsupervised fashion. A key challenge in this field is achieving robust and accurate depth estimation in challenging scenarios, particularly in regions with weak textures or where dynamic objects are present. This study makes three major contributions by delving deeply into dense correspondence priors to provide existing frameworks with explicit geometric constraints. The first novelty is a contextual-geometric depth consistency loss, which employs depth maps triangulated from dense correspondences based on estimated ego-motion to guide the learning of depth perception from contextual information, since explicitly triangulated depth maps capture accurate relative distances among pixels. The second novelty arises from the observation that there exists an explicit, deducible relationship between optical flow divergence and depth gradient. A differential property correlation loss is, therefore, designed to refine depth estimation with a specific emphasis on local variations. The third novelty is a bidirectional stream co-adjustment strategy that enhances the interaction between rigid and optical flows, encouraging the former towards more accurate correspondence and making the latter more adaptable across various scenarios under the static scene hypotheses. DCPI-Depth, a framework that incorporates all these innovative components and couples two bidirectional and collaborative streams, achieves state-of-the-art performance and generalizability across multiple public datasets, outperforming all existing prior arts. Specifically, it demonstrates accurate depth estimation in texture-less and dynamic regions, and shows more reasonable smoothness. Our source code will be publicly available at mias.group/DCPI-Depth upon publication.