PhysDepth: Plug-and-Play Physical Refinement for Monocular Depth Estimation in Challenging Environments

TL;DR

PhysDepth introduces a plug-and-play framework that enhances monocular depth estimation in challenging environments by integrating physical priors, specifically Rayleigh Scattering and Beer-Lambert law, leading to state-of-the-art performance.

Contribution

The paper presents PhysDepth, a novel physical prior-based framework that improves robustness of monocular depth estimation models in adverse conditions.

Findings

PhysDepth outperforms existing models in challenging environments.

Incorporating physical priors reduces error related to atmospheric attenuation.

The framework is plug-and-play and compatible with various backbones.

Abstract

State-of-the-art monocular depth estimation (MDE) models often struggle in challenging environments, primarily because they overlook robust physical information. To demonstrate this, we first conduct an empirical study by computing the covariance between a model's prediction error and atmospheric attenuation. We find that the error of existing SOTAs increases with atmospheric attenuation. Based on this finding, we propose PhysDepth, a plug-and-play framework that solves this fragility by infusing physical priors into modern SOTA backbones. PhysDepth incorporates two key components: a Physical Prior Module (PPM) that leverages Rayleigh Scattering theory to extract robust features from the high-SNR red channel, and a physics-derived Red Channel Attenuation Loss (RCA) that enforces model to learn the Beer-Lambert law. Extensive evaluations demonstrate that PhysDepth achieves SOTA accuracy in challenging conditions.