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

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 complex scene observation, we propose a ``Focus--Sweep'' paradigm and develop a Hierarchical Mamba 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.

Many applications, such as robots, 3D mapping, and visual localization, need to align a camera image with a 3D point cloud of the same scene. However, this is difficult because images and point clouds look very different, and changes in viewpoint or object scale can lead to wrong matches. We propose FS-I2P, a method that first looks at the whole scene to find rough connections, and then checks smaller regions to improve the matches. It also learns how many checking steps are needed instead of using a fixed number. Experiments show that our method gives more accurate and reliable alignment than previous approaches.