RoDyn-SLAM: Robust Dynamic Dense RGB-D SLAM with Neural Radiance Fields

TL;DR

RoDyn-SLAM introduces a neural radiance field-based dynamic RGB-D SLAM framework that effectively handles dynamic environments by filtering invalid rays, improving pose estimation, and achieving state-of-the-art accuracy and robustness.

Contribution

It presents a novel dynamic SLAM method using neural radiance fields with motion masks and a divide-and-conquer pose optimization for better performance in dynamic scenes.

Findings

Achieves state-of-the-art accuracy in dynamic environments

Demonstrates robustness across challenging datasets

Enhances geometry constraints with edge warp loss

Abstract

Leveraging neural implicit representation to conduct dense RGB-D SLAM has been studied in recent years. However, this approach relies on a static environment assumption and does not work robustly within a dynamic environment due to the inconsistent observation of geometry and photometry. To address the challenges presented in dynamic environments, we propose a novel dynamic SLAM framework with neural radiance field. Specifically, we introduce a motion mask generation method to filter out the invalid sampled rays. This design effectively fuses the optical flow mask and semantic mask to enhance the precision of motion mask. To further improve the accuracy of pose estimation, we have designed a divide-and-conquer pose optimization algorithm that distinguishes between keyframes and non-keyframes. The proposed edge warp loss can effectively enhance the geometry constraints between adjacent frames. Extensive experiments are conducted on the two challenging datasets, and the results show that RoDyn-SLAM achieves state-of-the-art performance among recent neural RGB-D methods in both accuracy and robustness.