FS-I2P:A Hierarchical Focus-Sweep Registration Network with Dynamically Allocated Depth

2026-05-08

AI Summary

This work addresses the challenges of scale ambiguity and erroneous correspondences in image-to-point-cloud registration caused by viewpoint variations, cross-modal discrepancies, and repetitive textures. To this end, the authors propose a “focus-and-scan” registration paradigm that leverages a state-space model to construct a hierarchical interaction module. Inspired by human visual perception, the method introduces a dynamic layer allocation strategy that adaptively determines the optimal iteration depth, thereby strengthening multi-scale cross-modal feature associations while mitigating attention drift and scale inconsistency. Evaluated on the RGB-D Scenes V2 and 7-Scenes benchmarks, the proposed approach achieves state-of-the-art performance, demonstrating significant improvements in both registration accuracy and robustness.

Abstract

Image-to-point cloud registration is often challenged by viewpoint changes, cross-modal discrepancies, and repetitive textures, which induce scale ambiguity and consequently lead to erroneous correspondences. Recent detection-free methods alleviate this issue by leveraging multi-scale features and transformer-based interactions. However, they still suffer from attention drift across layers and intra-scale inconsistencies, hindering precise registration. Inspired by human behavior, we propose a ``Focus--Sweep'' paradigm and develop a Hierarchical Focus--Sweep Interaction Module within an SSM-based framework to enhance multi-level cross-modal feature association. In addition, we introduce a Dynamic Layer Allocation Strategy that adaptively determines the iteration depth to better exploit geometric constraints and improve matching robustness. Extensive experiments and ablations on two benchmarks, RGB-D Scenes V2 and 7-Scenes, demonstrate that our approach achieves state-of-the-art performance.

Problem

• image-to-point cloud registration
• scale ambiguity
• cross-modal discrepancy
• attention drift
• intra-scale inconsistency

Innovation

• Focus-Sweep
• Hierarchical Interaction
• Dynamic Layer Allocation
• Image-to-Point Cloud Registration
• State Space Model