MSSP: Multi-Scale Spatially-Constrained Partition for Unsupervised Semantic Segmentation of 3D Point Clouds
A regularize-then-enrich interaction is revealed: multi-scale features alone do not improve final segmentation, yet become highly effective when coupled with spatial regularization, underscoring that spatial coherence is aprerequisite for multi-scale representations in superpoint clustering.
Zheng-Hao Zhang, Xin-Jie Wang, Wei Wang et al.
· 0 citations