SynWTS: Multi-View Digital-Twin Dataset for Synthetic-to-Real Traffic Safety Understanding

• Paper

Summary

SynWTS pairs synthetic digital-twin videos with real recordings from the Woven Traffic Safety dataset. It contains 249 paired scenarios and 846 synchronized videos from overhead and ego-vehicle perspectives, with bounding boxes, textual descriptions, visual question-answer pairs, and accident phases. This pairing supports direct study of synthetic-to-real transfer and the effect of visual realism on traffic-safety reasoning.

Citation

@inproceedings{Patil26SynWTS,
  author = {Dhanishtha Patil and Ridham Kachhadiya and Andrew Vattuone and Haoquan Liang and Jiajun Li and Yuxing Wang and Zheng Tang and Ashutosh Kumar and Quan Kong and David C. Anastasiu},
  title = { {SynWTS}: {M}ulti-view digital-twin dataset for synthetic-to-real traffic safety understanding},
  booktitle = {Proc. ECCV Workshops},
  address = {Malm{\"o}, Sweden},
  year = {2026}
}