Research map / Point clouds / Registration and matching

Register Anything Model for Generalizable and Robust Point Cloud Registration

NeurIPS 2026 · Point clouds › Registration and matching · approach: Learned feature-based registration

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Register Anything Model (RAM), a unified model for generalizable 3D point cloud registration across diverse sensors and environments, aiming to avoid per-domain tuning.

Problem

Existing point cloud registration methods are tailored to a single domain with carefully tuned architectures and hyperparameters, and degrade markedly on novel domains.

Approach

A single unified model trained to estimate 6-DoF transformations across diverse domains (sensors and environments), intended to generalize to novel domains without domain-specific tuning. Abstract is truncated, so the specific training mechanism is not stated.

Closest papers by meaning

• Generalized-CVO: Fast and Correspondence-Free Local Point Cloud Registration with Second Order Riemannian Optimization (CVPR 2026 · similarity 0.90)
• FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.90)
• Mining and Transferring Feature-Geometry Coherence for Unsupervised Point Cloud Registration (NeurIPS 2024 · similarity 0.89)
• HybridReg: Robust 3D Point Cloud Registration with Hybrid Motions (AAAI 2025 · similarity 0.88)
• Where Precision Meets Efficiency: Transformation Diffusion Model for Point Cloud Registration (AAAI 2025 · similarity 0.88)
• A robust inlier identification algorithm for point cloud registration via $\mathbf{\ell_0}$-minimization (NeurIPS 2024 · similarity 0.88)