Exact Point Cloud Downsampling for Fast and Accurate Global Trajectory Optimization

TL;DR

This paper introduces an exact point cloud downsampling method that significantly speeds up and reduces memory use in trajectory optimization tasks without sacrificing accuracy by selecting a minimal residual subset.

Contribution

The proposed algorithm precisely preserves the registration error function using a small residual subset, enabling faster and more memory-efficient global registration.

Findings

Reduces processing time by 87%

Decreases memory consumption by 99%

Maintains accuracy in trajectory optimization

Abstract

This paper presents a point cloud downsampling algorithm for fast and accurate trajectory optimization based on global registration error minimization. The proposed algorithm selects a weighted subset of residuals of the input point cloud such that the subset yields exactly the same quadratic point cloud registration error function as that of the original point cloud at the evaluation point. This method accurately approximates the original registration error function with only a small subset of input points (29 residuals at a minimum). Experimental results using the KITTI dataset demonstrate that the proposed algorithm significantly reduces processing time (by 87\%) and memory consumption (by 99\%) for global registration error minimization while retaining accuracy.