PC2Model: ISPRS benchmark on 3D point cloud to model registration

arXiv cs.CV / 4/22/2026

📰 NewsDeveloper Stack & InfrastructureModels & Research

共有:

Key Points

The paper focuses on PC2Model (point cloud-to-3D model) registration, a key step for aligning multimodal 3D data in areas like construction monitoring, autonomous driving, robotics, and VR/AR.
It argues that data-driven registration methods often underperform on real scans due to sparsity, noise, clutter, and occlusions.
To address these gaps, the authors introduce the PC2Model benchmark as a public dataset for training and evaluating both classical and learning-based methods.
The benchmark uses a hybrid design combining simulated point clouds (with precise ground truth) and, in some cases, real-world scans paired with corresponding 3D models (to include sensor/environment artifacts).
The setup is intended to enable robust cross-domain evaluation and systematic study of how well models trained on simulation transfer to real-world scenarios.

Abstract

Point cloud registration involves aligning one point cloud with another or with a three-dimensional (3D) model, enabling the integration of multimodal data into a unified representation. This is essential in applications such as construction monitoring, autonomous driving, robotics, and virtual or augmented reality (VR/AR).With the increasing accessibility of point cloud acquisition technologies, such as Light Detection and Ranging (LiDAR) and structured light scanning, along with recent advances in deep learning, the research focus has increasingly shifted towards downstream tasks, particularly point cloud-to-model (PC2Model) registration. While data-driven methods aim to automate this process, they struggle with sparsity, noise, clutter, and occlusions in real-world scans, which limit their performance. To address these challenges, this paper introduces the PC2Model benchmark, a publicly available dataset designed to support the training and evaluation of both classical and data-driven methods. Developed under the leadership of ICWG II/Ib, the PC2Model benchmark adopts a hybrid design that combines simulated point clouds with, in some cases, real-world scans and their corresponding 3D models. Simulated data provide precise ground truth and controlled conditions, while real-world data introduce sensor and environmental artefacts. This design supports robust training and evaluation across domains and enables the systematic analysis of model transferability from simulated to real-world scenarios. The dataset is publicly accessible at: https://zenodo.org/uploads/17581812.