发表机构
Department of Computer Science and Engineering, Toyohashi University of Technology; National Institute of Advanced Industrial Science and Technology（丰桥技术科学大学计算机科学与工程系; 国立先进工业科学技术研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究图像到点云配准难题，提出将激光雷达视为成像传感器的方法，通过条件整流流、特征匹配等步骤估计位姿，经自监督预训练和少量数据微调，实现高精度快速配准，性能优于现有方法。
AI 中文摘要
图像到点云配准（I2P）对集成相机和激光雷达至关重要，但模态差异使其难以兼顾高精度与强泛化性。本文提出一种简单有效的I2P方法，将激光雷达视为成像传感器，用条件整流流生成密集激光雷达强度图像，通过预训练特征匹配器与相机图像匹配，经PnP-RANSAC估计6自由度相对位姿。模型经自监督图像完成任务预训练，在少量激光雷达数据上微调，能适应多种配置。实验表明该方法平均误差4.89° / 1.63 m，优于现有方法，单次配准约0.68 s。
英文摘要
Image-to-Point Cloud Registration (I2P) is essential for integrating camera and LiDAR in perception and autonomous systems, yet the modality gap between images and point clouds makes it difficult to achieve both high accuracy and strong generalization. In this paper, we propose a simple yet effective I2P method that treats LiDAR as an imaging sensor: from a single sparse LiDAR scan, we generate a dense LiDAR intensity image using Conditional Rectified Flow, match it with a camera image using a pre-trained feature matcher, and estimate the 6-DoF relative pose via PnP-RANSAC. The proposed model is pre-trained through a self-supervised image completion task and fine-tuned on a small amount of LiDAR data (neither image-point cloud pairs nor ground-truth sensor poses are required), enabling it to scale to diverse LiDAR and camera configurations. Experiments on the R3LIVE dataset show that the proposed method achieves a mean error of 4.89° / 1.63 m, outperforming existing methods, while completing a single registration in approximately 0.68 s.
