Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces R-SLPR, a novel framework for point-cloud registration that addresses the challenge of aligning small, partial point clouds with larger reference clouds. By reformulating the registration process into three stages—region proposal, regional matching, and iterative refinement—R-SLPR enhances robustness against scale mismatches, a common issue in real-world applications. The framework achieves state-of-the-art accuracy, significantly reducing mean absolute errors in position and rotation compared to existing methods.
R-SLPR achieves unprecedented accuracy in point-cloud registration, reducing position and rotation errors to 0.009 and 1.104, respectively, even under severe scale mismatches.
Point-cloud (PC) registration is fundamental to three-dimensional (3D) perception in robotic systems. However, classic registration algorithms falter when aligning a source PC containing limited, incomplete, or ambiguous geometric cues against a reference. This challenge of registering a small, partial PC to a significantly larger global reference is pervasive in real-world deployment yet remains insufficiently addressed by existing learning-based approaches, which typically assume comparable scales and significant overlap. To bridge this gap, we propose the Region-based Small-to-Large Point-cloud Registra- tion framework (R-SLPR), a novel three-stage architecture that fundamentally reformulates the scale-mismatched registration problem into a sequence of region proposal, regional matching, and iterative refinement. Unlike conventional methods that fail to localize specific regions, R-SLPR explicitly identifies candidate regions prior to estimating rigid transformations, ensuring robust alignment even under severe scale mismatch. The framework introduces a Fibonacci Grid Segmentation method coupled with a contrastive learning objective to effectively generate and match local geometric patches. Building on this, a novel Cascade Anchor Selection and Refinement algorithm iteratively aligns the source with the target region to maximize precision. Extensive evaluation on ModelNet40 demonstrates that R-SLPR establishes a new state-of-the-art accuracy standard, outperforming prior approaches and significantly reducing position and rotation Mean Absolute Error (MAE) to 0.009 and 1.104, respectively.