--- tags: - traffic language: - en task_categories: - question-answering license: other --- # AV-TAU Dataset ## Overview The **AV-TAU Dataset** is developed for advancing research in traffic anomaly understanding and reasoning. It provides carefully annotated question–answer pairs in English, aligned with corresponding video data, and is intended **exclusively for academic and educational use**. --- ## Related Publication 🔗 [EchoTraffic: Enhancing Traffic Anomaly Understanding with Audio-Visual Insights (CVPR 2025)](https://openaccess.thecvf.com/content/CVPR2025/papers/Xing_EchoTraffic_Enhancing_Traffic_Anomaly_Understanding_with_Audio-Visual_Insights_CVPR_2025_paper.pdf) --- ## ⚠️ Usage Restrictions (Academic Use Only) Access to this dataset is **gated**. To obtain access, researchers must complete the following steps: 1. **Download and sign the Academic Use License Agreement** - The full license text is provided in the [LICENSE](https://github.com/HarryHsing/EchoTraffic/blob/main/LICENSE.pdf) file. - Print or digitally sign the document. 2. **Send the signed agreement** - Email the signed copy to: 📧 harryhsing@outlook.com 3. **Wait for approval** - Your Hugging Face access request will only be approved once we confirm receipt of your signed agreement. - Requests without a signed agreement will be rejected. --- ## Citation If you use this dataset in your research, please cite it as: ```bibtex @InProceedings{Xing_2025_CVPR, author = {Xing, Zhenghao and Chen, Hao and Xie, Binzhu and Xu, Jiaqi and Guo, Ziyu and Xu, Xuemiao and Hao, Jianye and Fu, Chi-Wing and Hu, Xiaowei and Heng, Pheng-Ann}, title = {EchoTraffic: Enhancing Traffic Anomaly Understanding with Audio-Visual Insights}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2025}, pages = {19098-19108} } ``` --- ## Contact For questions or special access requests, please contact: 📧 harryhsing@outlook.com