LATITUDE: Robotic Global Localization with Truncated Dynamic Low-pass Filter in City-scale NeRF

TL;DR

LATITUDE is a two-stage city-scale NeRF-based localization method that combines a regressor for initial pose estimation with a TDLF-enhanced optimization for high-precision global localization.

Contribution

It introduces a novel two-stage localization framework with a Truncated Dynamic Low-pass Filter to improve global pose estimation in large-scale city scenes.

Findings

Effective in synthetic and real-world datasets

Achieves high-precision localization in city-scale environments

Outperforms existing NeRF-based localization methods

Abstract

Neural Radiance Fields (NeRFs) have made great success in representing complex 3D scenes with high-resolution details and efficient memory. Nevertheless, current NeRF-based pose estimators have no initial pose prediction and are prone to local optima during optimization. In this paper, we present LATITUDE: Global Localization with Truncated Dynamic Low-pass Filter, which introduces a two-stage localization mechanism in city-scale NeRF. In place recognition stage, we train a regressor through images generated from trained NeRFs, which provides an initial value for global localization. In pose optimization stage, we minimize the residual between the observed image and rendered image by directly optimizing the pose on tangent plane. To avoid convergence to local optimum, we introduce a Truncated Dynamic Low-pass Filter (TDLF) for coarse-to-fine pose registration. We evaluate our method on both synthetic and real-world data and show its potential applications for high-precision navigation in large-scale city scenes. Codes and data will be publicly available at https://github.com/jike5/LATITUDE.