Towards Zero-Shot Point Cloud Registration Across Diverse Scales, Scenes, and Sensor Setups
Hyungtae Lim, Minkyun Seo, Luca Carlone, Jaesik Park

TL;DR
BUFFER-X is a training-free, zero-shot point cloud registration framework that automatically adapts to diverse scales, scenes, and sensor setups, outperforming existing methods without domain-specific tuning.
Contribution
We introduce BUFFER-X, a novel registration method that eliminates the need for training and manual hyperparameter tuning, enabling robust zero-shot performance across varied environments.
Findings
Achieves 43% faster computation with BUFFER-X-Lite.
Generalizes effectively across 12 diverse datasets.
Outperforms existing methods without domain-specific tuning.
Abstract
Some deep learning-based point cloud registration methods struggle with zero-shot generalization, often requiring dataset-specific hyperparameter tuning or retraining for new environments. We identify three critical limitations: (a) fixed user-defined parameters (e.g., voxel size, search radius) that fail to generalize across varying scales, (b) learned keypoint detectors exhibit poor cross-domain transferability, and (c) absolute coordinates amplify scale mismatches between datasets. To address these three issues, we present BUFFER-X, a training-free registration framework that achieves zero-shot generalization through: (a) geometric bootstrapping for automatic hyperparameter estimation, (b) distribution-aware farthest point sampling to replace learned detectors, and (c) patch-level coordinate normalization to ensure scale consistency. Our approach employs hierarchical multi-scale…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRobotics and Sensor-Based Localization · 3D Shape Modeling and Analysis · Robot Manipulation and Learning
