RobustLoc: Robust Camera Pose Regression in Challenging Driving Environments

TL;DR

RobustLoc introduces a neural differential equation-based approach for camera pose regression that maintains high accuracy across diverse challenging driving environments with environmental perturbations.

Contribution

It presents a novel neural differential equation diffusion module and a branched pose decoder to improve robustness in camera relocalization under environmental variability.

Findings

Outperforms existing camera pose regression models in diverse conditions

Demonstrates robustness against weather, lighting, and seasonal changes

Achieves state-of-the-art accuracy in challenging driving scenarios

Abstract

Camera relocalization has various applications in autonomous driving. Previous camera pose regression models consider only ideal scenarios where there is little environmental perturbation. To deal with challenging driving environments that may have changing seasons, weather, illumination, and the presence of unstable objects, we propose RobustLoc, which derives its robustness against perturbations from neural differential equations. Our model uses a convolutional neural network to extract feature maps from multi-view images, a robust neural differential equation diffusion block module to diffuse information interactively, and a branched pose decoder with multi-layer training to estimate the vehicle poses. Experiments demonstrate that RobustLoc surpasses current state-of-the-art camera pose regression models and achieves robust performance in various environments. Our code is released at: https://github.com/sijieaaa/RobustLoc