SPEAL: Skeletal Prior Embedded Attention Learning for Cross-Source Point Cloud Registration

arXiv:2312.08664v2 Announce Type: replace
Abstract: Point cloud registration, a fundamental task in 3D computer vision, has remained largely unexplored in cross-source point clouds and unstructured scenes. The primary challenges arise from noise, outliers, and variations in scale and density. However, neglected geometric natures of point clouds restricts the performance of current methods. In this paper, we propose a novel method termed SPEAL to leverage skeletal representations for effective learning of intrinsic topologies of point clouds, facilitating robust capture of geometric …

abstract arxiv attention challenges cloud computer computer vision cs.cv embedded noise outliers performance prior registration scale type unstructured vision