[IEEE TCSVT 2024] Neuromorphic Human Action Recognition Dataset (NeuroHAR)

The NeuroHAR dataset is constructed by the Robot Vision Research Laboratory at City University of Hong Kong. For more details, please read our IEEE TCSVT paper 'Event Voxel Set Transformer for Spatiotemporal Representation Learning on Event Streams'.

If you use this dataset for experiments, please cite the following publication:

@article{xie2024evstr,
  title={Event Voxel Set Transformer for Spatiotemporal Representation Learning on Event Streams},
  author={Xie, Bochen and Deng, Yongjian and Shao, Zhanpeng and Xu, Qingsong and Li, Youfu},
  journal={IEEE Trans. Circuits Syst. Video Technol.},
  pages={1--14},
  year={2024},
  note = {doi: \href{https://dx.doi.org/10.1109/TCSVT.2024.3448615}{\textcolor{blue}{10.1109/TCSVT.2024.3448615}}}
}

Download

• Event Data: [OneDrive]
• RGB Images: to be uploaded
• Depth Images: to be uploaded