Enhanced Low-resolution LiDAR-Camera Calibration Via Depth Interpolation and Supervised Contrastive Learning

TL;DR

This paper introduces a novel calibration method for low-resolution LiDAR-camera systems that uses depth interpolation and supervised contrastive learning to improve accuracy despite sparsity and noise.

Contribution

The paper proposes a new calibration approach combining depth interpolation and supervised contrastive learning specifically for low-resolution LiDAR data.

Findings

Achieves 0.15cm mean absolute rotation error

Achieves 0.33° mean absolute translation error

Outperforms existing calibration methods on RELLIS-3D dataset

Abstract

Motivated by the increasing application of low-resolution LiDAR recently, we target the problem of low-resolution LiDAR-camera calibration in this work. The main challenges are two-fold: sparsity and noise in point clouds. To address the problem, we propose to apply depth interpolation to increase the point density and supervised contrastive learning to learn noise-resistant features. The experiments on RELLIS-3D demonstrate that our approach achieves an average mean absolute rotation/translation errors of 0.15cm/0.33\textdegree on 32-channel LiDAR point cloud data, which significantly outperforms all reference methods.