Robust Geometry-Preserving Depth Estimation Using Differentiable Rendering

TL;DR

This paper introduces a novel learning framework for monocular depth estimation that preserves geometric consistency across views, enabling accurate 3D scene reconstruction without extra data or annotations.

Contribution

It presents a geometry-preserving depth estimation method that generalizes well across datasets and recovers scale and shift factors using only unlabeled images.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Does not require additional 3D data or annotations

Automatically recovers scale-and-shift coefficients

Abstract

In this study, we address the challenge of 3D scene structure recovery from monocular depth estimation. While traditional depth estimation methods leverage labeled datasets to directly predict absolute depth, recent advancements advocate for mix-dataset training, enhancing generalization across diverse scenes. However, such mixed dataset training yields depth predictions only up to an unknown scale and shift, hindering accurate 3D reconstructions. Existing solutions necessitate extra 3D datasets or geometry-complete depth annotations, constraints that limit their versatility. In this paper, we propose a learning framework that trains models to predict geometry-preserving depth without requiring extra data or annotations. To produce realistic 3D structures, we render novel views of the reconstructed scenes and design loss functions to promote depth estimation consistency across different views. Comprehensive experiments underscore our framework's superior generalization capabilities, surpassing existing state-of-the-art methods on several benchmark datasets without leveraging extra training information. Moreover, our innovative loss functions empower the model to autonomously recover domain-specific scale-and-shift coefficients using solely unlabeled images.