This repository contains the official implementation of the paper:
Mamba3D: Enhancing Local Features for 3D Point Cloud Analysis via State Space Model
- We present Mamba3D, a state space model tailored for point cloud learning.
- Mamba3D surpasses Transformer-based counterparts and concurrent works in multiple tasks, achieving multiple SoTA, with linear complexity.
- Release the training and evaluation code
- Release the pretrained weights
- Release the toy code on Colab
We would like to thank the authors of Mamba, Vision Mamba, and Point-MAE for their great works and repos.
If you find our work helpful, please consider citing:
@article{han2024mamba3d,
title={Mamba3D: Enhancing Local Features for 3D Point Cloud Analysis via State Space Model},
author={Han, Xu and Tang, Yuan and Wang, Zhaoxuan and Li, Xianzhi},
journal={arXiv preprint arXiv:2404.14966},
year={2024}
}