README.md

metadata

license: mit
library_name: torch-pointcloud
tags:
  - point-cloud
  - 3d
  - pytorch
  - torch-pointcloud
  - pointnet2
  - classification
datasets:
  - scanobjectnn
model-index:
  - name: pointnet2.scanobjectnn-hardest.openpoints
    results:
      - task:
          type: point-cloud-classification
        dataset:
          name: ScanObjectNN (PB_T50_RS)
          type: scanobjectnn
        metrics:
          - name: OA
            type: accuracy
            value: 86.16
          - name: mAcc
            type: accuracy
            value: 84.36

Model card for pointnet2.scanobjectnn-hardest.openpoints

A PointNet++ point cloud classification model (hierarchical set abstraction). Trained on ScanObjectNN (PB_T50_RS).

Model Details

• Model Type: Point cloud classification
• Model Stats:
  • Params (M): 1.5
  • Input channels: 4
  • Classes: 15
  • Features: 1024
• Dataset: ScanObjectNN (PB_T50_RS)
• Metrics: OA 86.16, mAcc 84.36 (reference 86.2)
• Paper: PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
• Converted from: guochengqian/PointNeXt (MIT)
• Library: torch-pointcloud

Install

pip install torch-pointcloud

Usage

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

model, info = tp.create_model(
    "pointnet2.scanobjectnn-hardest.openpoints",
    pretrained=True,
    return_info=True,
)
model = model.eval()

# 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])

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

Feature extraction

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, 1024)

Citation

@inproceedings{qi2017pointnet2,
  title   = {PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space},
  author  = {Charles R. Qi and Li Yi and Hao Su and Leonidas J. Guibas},
  booktitle = {NeurIPS},
  year    = {2017}
}

@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},
}