E-Mamba: An efficient Mamba point cloud analysis method with enhanced feature representation

Dengao Li, Zhichao Gao, Shufeng Hao, Ziyou Xun, Jiajian Song, Jie Cheng, Jumin Zhao

Abstract: Highlights•The point cloud density varies greatly between different samples.•Scalable Embedding improves the model’s adaptability to non-uniform points.•Holosync Reordering Pooling improves global feature extraction capabilities.•The E-Mamba model significantly improves classification accuracy and robustness.