Dusk Till Dawn: Self-supervised Nighttime Stereo Depth Estimation using Visual Foundation Models

TL;DR

This paper introduces a self-supervised stereo depth estimation method tailored for nighttime conditions, leveraging pretrained visual foundation models and novel masking and evaluation techniques to improve accuracy in challenging low-light scenarios.

Contribution

We propose a new approach that uses visual foundation models for feature extraction, a masking strategy to handle photometric violations, and novel evaluation metrics for nighttime stereo depth estimation.

Findings

Significant improvement on Oxford RobotCar and Multi-Spectral Stereo datasets.

Robust depth estimation in low-visibility and nighttime conditions.

Enhanced evaluation metrics for depth accuracy.

Abstract

Self-supervised depth estimation algorithms rely heavily on frame-warping relationships, exhibiting substantial performance degradation when applied in challenging circumstances, such as low-visibility and nighttime scenarios with varying illumination conditions. Addressing this challenge, we introduce an algorithm designed to achieve accurate self-supervised stereo depth estimation focusing on nighttime conditions. Specifically, we use pretrained visual foundation models to extract generalised features across challenging scenes and present an efficient method for matching and integrating these features from stereo frames. Moreover, to prevent pixels violating photometric consistency assumption from negatively affecting the depth predictions, we propose a novel masking approach designed to filter out such pixels. Lastly, addressing weaknesses in the evaluation of current depth estimation algorithms, we present novel evaluation metrics. Our experiments, conducted on challenging datasets including Oxford RobotCar and Multi-Spectral Stereo, demonstrate the robust improvements realized by our approach. Code is available at: https://github.com/madhubabuv/dtd