Model card for spformer.scannet.jiahao-sun

A SPFormer point cloud instance segmentation model (superpoint transformer). Trained on ScanNet.

Model Details

• Model Type: Point cloud instance segmentation
• Model Stats:
  • Params (M): 17.6
  • Input channels: 6
  • Classes: 18
  • Features: 32
• Dataset: ScanNet
• Metrics: mAP 56.47, mAP@0.5 73.93, mAP@0.25 82.55
• Paper: Superpoint Transformer for 3D Scene Instance Segmentation
• Converted from: sunjiahao1999/SPFormer (MIT)
• Library: torch-pointcloud

Install

pip install torch-pointcloud

This checkpoint also needs spconv, which needs a build matching your torch and CUDA: see the installation guide.

Usage

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

model, info = tp.create_model(
    "spformer.scannet.jiahao-sun",
    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),
    "color": torch.rand(num_points, 3) * 255,
    "segment": torch.zeros(num_points, dtype=torch.long),
    "superpoint": torch.randint(0, 64, (num_points,)),
}
data = info["transform"](sample)
data = collate([data], cat_keys=["inverse"])
data = {key: value.cuda() for key, value in data.items()}

with torch.no_grad():
    out = model(
        data.get("x"),
        data["pos_grid"],
        data["batch"],
        data["superpoint"],
        data["inverse"],
        data["batch_inverse"],
    )
    instances = model.decode(out)  # mask_index (2, M), labels, scores and batch (K,)

Feature extraction

with torch.no_grad():
    features, superpoint, superpoint_batch = model.forward_features(
        data.get("x"),
        data["pos_grid"],
        data["batch"],
        data["superpoint"],
        data["inverse"],
        data["batch_inverse"],
    )  # 32 channels per superpoint

Citation

@inproceedings{sun2023spformer,
  title   = {Superpoint Transformer for 3D Scene Instance Segmentation},
  author  = {Jiahao Sun and Chunmei Qing and Junpeng Tan and Xiangmin Xu},
  booktitle = {AAAI},
  year    = {2023}
}

@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},
}