README.md

metadata

license: mit
library_name: torch-pointcloud
tags:
  - point-cloud
  - 3d
  - pytorch
  - torch-pointcloud
  - votenet
  - object-detection
datasets:
  - scannet
model-index:
  - name: votenet.scannet.fair
    results:
      - task:
          type: 3d-object-detection
        dataset:
          name: ScanNet
          type: scannet
        metrics:
          - name: mAP@0.25
            type: map
            value: 57.65
          - name: mAP@0.5
            type: map
            value: 34.1

Model card for votenet.scannet.fair

A VoteNet 3D object detection model (deep Hough voting detector). Trained on ScanNet.

Model Details

• Model Type: 3D object detection
• Model Stats:
  • Params (M): 1.0
  • Input channels: 1
  • Classes: 18
  • Features: 256
• Dataset: ScanNet
• Metrics: mAP@0.25 57.65, mAP@0.5 34.1 (reference 58.6)
• Paper: Deep Hough Voting for 3D Object Detection in Point Clouds
• Converted from: facebookresearch/votenet (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(
    "votenet.scannet.fair",
    task="detection",
    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),
    "color": torch.rand(num_points, 3) * 255,
    "normal": torch.randn(num_points, 3),
    "segment": torch.zeros(num_points, dtype=torch.long),
    "instance": torch.zeros(num_points, dtype=torch.long),
}
data = info["transform"](sample)
data = collate([data])

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

Feature extraction

with torch.no_grad():
    features = model.forward_features(data.get("x"), data["pos"], data["batch"])  # 256 channels

Citation

@inproceedings{qi2019votenet,
  title   = {Deep Hough Voting for 3D Object Detection in Point Clouds},
  author  = {Charles R. Qi and Or Litany and Kaiming He and Leonidas J. Guibas},
  booktitle = {ICCV},
  year    = {2019}
}

@inproceedings{dai2017scannet,
  title   = {ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes},
  author  = {Angela Dai and Angel X. Chang and Manolis Savva and Maciej Halber and Thomas Funkhouser and Matthias Nießner},
  booktitle = {CVPR},
  year    = {2017}
}

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