Attention-guided reference point shifting for Gaussian-mixture-based partial point set registration
This addresses a domain-specific problem in point set registration for applications like 3D vision, but it is incremental as it builds on existing methods like DeepGMR.
The study tackled the problem of partial-to-partial point set registration by identifying issues with invariance in deep learning and Gaussian mixture model methods, and proposed an attention-based reference point shifting layer that significantly enhanced the performance of DeepGMR and UGMMReg, outperforming prior deep learning methods.
This study investigates the impact of the invariance of feature vectors for partial-to-partial point set registration under translation and rotation of input point sets, particularly in the realm of techniques based on deep learning and Gaussian mixture models (GMMs). We reveal both theoretical and practical problems associated with such deep-learning-based registration methods using GMMs, with a particular focus on the limitations of DeepGMR, a pioneering study in this line, to the partial-to-partial point set registration. Our primary goal is to uncover the causes behind such methods and propose a comprehensible solution for that. To address this, we introduce an attention-based reference point shifting (ARPS) layer, which robustly identifies a common reference point of two partial point sets, thereby acquiring transformation-invariant features. The ARPS layer employs a well-studied attention module to find a common reference point rather than the overlap region. Owing to this, it significantly enhances the performance of DeepGMR and its recent variant, UGMMReg. Furthermore, these extension models outperform even prior deep learning methods using attention blocks and Transformer to extract the overlap region or common reference points. We believe these findings provide deeper insights into registration methods using deep learning and GMMs.