Point Cloud Registration for LiDAR and Photogrammetric Data: a Critical Synthesis and Performance Analysis on Classic and Deep Learning Algorithms

TL;DR

This paper critically reviews and evaluates over ten point cloud registration algorithms, including classic and deep learning methods, across diverse datasets, revealing significant performance gaps and guiding future research directions.

Contribution

It provides a comprehensive, two-step analysis of state-of-the-art registration methods across various data sources, highlighting their strengths, limitations, and areas for improvement.

Findings

Most algorithms have success rates below 40% on tested datasets.

Deep learning methods show potential but still lag behind classic algorithms.

There is a large margin for improvement in 3D correspondence search and complex scene registration.

Abstract

Recent advances in computer vision and deep learning have shown promising performance in estimating rigid/similarity transformation between unregistered point clouds of complex objects and scenes. However, their performances are mostly evaluated using a limited number of datasets from a single sensor (e.g. Kinect or RealSense cameras), lacking a comprehensive overview of their applicability in photogrammetric 3D mapping scenarios. In this work, we provide a comprehensive review of the state-of-the-art (SOTA) point cloud registration methods, where we analyze and evaluate these methods using a diverse set of point cloud data from indoor to satellite sources. The quantitative analysis allows for exploring the strengths, applicability, challenges, and future trends of these methods. In contrast to existing analysis works that introduce point cloud registration as a holistic process, our experimental analysis is based on its inherent two-step process to better comprehend these approaches including feature/keypoint-based initial coarse registration and dense fine registration through cloud-to-cloud (C2C) optimization. More than ten methods, including classic hand-crafted, deep-learning-based feature correspondence, and robust C2C methods were tested. We observed that the success rate of most of the algorithms are fewer than 40% over the datasets we tested and there are still are large margin of improvement upon existing algorithms concerning 3D sparse corresopondence search, and the ability to register point clouds with complex geometry and occlusions. With the evaluated statistics on three datasets, we conclude the best-performing methods for each step and provide our recommendations, and outlook future efforts.