NuRF: Nudging the Particle Filter in Radiance Fields for Robot Visual Localization

TL;DR

NuRF introduces an adaptive particle filter framework leveraging radiance fields and novel view synthesis to improve indoor robot visual localization, achieving faster convergence and high accuracy using monocular vision.

Contribution

It presents a novel framework combining radiance fields with nudged particle filtering for improved monocular robot localization.

Findings

Converges 7 times faster than existing Monte Carlo methods.

Achieves localization accuracy within 1 meter.

Enhances visual localization with novel view synthesis.

Abstract

Can we localize a robot on a map only using monocular vision? This study presents NuRF, an adaptive and nudged particle filter framework in radiance fields for 6-DoF robot visual localization. NuRF leverages recent advancements in radiance fields and visual place recognition. Conventional visual place recognition meets the challenges of data sparsity and artifact-induced inaccuracies. By utilizing radiance field-generated novel views, NuRF enhances visual localization performance and combines coarse global localization with the fine-grained pose tracking of a particle filter, ensuring continuous and precise localization. Experimentally, our method converges 7 times faster than existing Monte Carlo-based methods and achieves localization accuracy within 1 meter, offering an efficient and resilient solution for indoor visual localization.