The Open-Vocabulary ScanNet datasets from CoDA and CoDAv2.
If the dataset is helpful, please cite:
@inproceedings{dai2017scannet,
title={ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes},
author={Dai, Angela and Chang, Angel X. and Savva, Manolis and Halber, Maciej and Funkhouser, Thomas and Nie{\ss}ner, Matthias},
booktitle = {Proc. Computer Vision and Pattern Recognition (CVPR), IEEE},
year = {2017}
}
@inproceedings{cao2023coda,
title={CoDA: Collaborative Novel Box Discovery and Cross-modal Alignment for Open-vocabulary 3D Object Detection},
author={Cao, Yang and Zeng, Yihan and Xu, Hang and Xu, Dan},
booktitle={NeurIPS},
year={2023}
}
@article{cao2024collaborative,
title={Collaborative Novel Object Discovery and Box-Guided Cross-Modal Alignment for Open-Vocabulary 3D Object Detection},
author={Cao, Yang and Zeng, Yihan and Xu, Hang and Xu, Dan},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
year={2025}
}