---
license: apache-2.0
library_name: torch-pointcloud
tags:
- point-cloud
- 3d
- pytorch
- torch-pointcloud
- point-mamba
- classification
datasets:
- scanobjectnn
base_model: torch-pointcloud/point-mamba-base.pretrain.dingkang-liang
model-index:
- name: point-mamba-base.scanobjectnn.dingkang-liang
  results:
  - task:
      type: point-cloud-classification
    dataset:
      name: ScanObjectNN
      type: scanobjectnn
    metrics:
    - name: OA
      type: accuracy
      value: 94.32
---

# Model card for point-mamba-base.scanobjectnn.dingkang-liang

A PointMamba point cloud classification model (state space model over serialized points). Trained on ScanObjectNN.

## Model Details

- **Model Type:** Point cloud classification
- **Model Stats:**
  - Params (M): 12.3
  - Classes: 15
  - Features: 384
- **Dataset:** ScanObjectNN
- **Metrics:** OA 94.32 (reference 94.32)
- **Paper:** [PointMamba: A Simple State Space Model for Point Cloud Analysis](https://arxiv.org/abs/2402.10739)
- **Converted from:** [LMD0311/PointMamba](https://github.com/LMD0311/PointMamba) (Apache-2.0)
- **Library:** [torch-pointcloud](https://github.com/arthurdjn/pytorch-pointcloud)

## Install

```bash
pip install torch-pointcloud
```

This checkpoint also needs `mamba-ssm`, which needs a build matching your torch and CUDA: see the [installation guide](https://pytorch-pointcloud.org/installation/).

## Usage

```python
import torch
import torch_pointcloud as tp
from torch_pointcloud.utils.data import collate

model, info = tp.create_model(
    "point-mamba-base.scanobjectnn.dingkang-liang",
    task="classification",
    pretrained=True,
    return_info=True,
)
model = model.cuda().eval()  # GPU-only kernels

# synthetic sample with the keys a dataset provides
num_points = 8192
sample = {
    "pos": torch.randn(num_points, 3),
}
data = info["transform"](sample)
data = collate([data])
data = {key: value.cuda() for key, value in data.items()}

with torch.no_grad():
    logits = model(data.get("x"), data["pos"], data["batch"])
```

## Feature extraction

```python
with torch.no_grad():
    embeddings = model.forward_features(data.get("x"), data["pos"], data["batch"])

model.reset_classifier(num_classes=0)
with torch.no_grad():
    embeddings = model(data.get("x"), data["pos"], data["batch"])  # (B, 384)
```

## Citation

```bibtex
@inproceedings{liang2024pointmamba,
  title   = {PointMamba: A Simple State Space Model for Point Cloud Analysis},
  author  = {Dingkang Liang and Xin Zhou and Wei Xu and Xingkui Zhu and Zhikang Zou and Xiaoqing Ye and Xiao Tan and Xiang Bai},
  booktitle = {NeurIPS},
  year    = {2024}
}

@inproceedings{uy2019scanobjectnn,
  title   = {Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data},
  author  = {Mikaela Angelina Uy and Quang-Hieu Pham and Binh-Son Hua and Duc Thanh Nguyen and Sai-Kit Yeung},
  booktitle = {ICCV},
  year    = {2019}
}

@software{dujardin2026pytorchpointcloud,
  author  = {Arthur Dujardin},
  title   = {PyTorch PointCloud},
  year    = {2026},
  doi     = {10.5281/zenodo.22159632},
  url     = {https://github.com/arthurdjn/pytorch-pointcloud},
}
```