Ready to stress-test your detector in the wild? The 6th RWS workshop introduce a new challenge on Robust Thermal-Image Object Detection to advance multi-object detection performance under long-term thermal drift.
Real-world thermal imaging is messy: sensors re-calibrate, ambient temperatures swing, and weather keeps changing, slowly eroding detector performance. The RWS Robust Thermal-Image Object Detection Challenge puts this problem front using the LTDv2, a large-scale dataset purpose-built for bechmarking long-term robustness wtih 1 million frames from video over 8 months, 6.8+ million annotated boxes, and rich weather metadata.
Your mission: build object detectors that don’t just perform well on a sunny day, but stay consistent across seasons, weather patterns, and day–night cycles. We’ll score submissions on overall mAP and temporal consistency to reward models that are both accurate and stable over time. The challenge features a development phase for method iteration and a final test phase for the leaderboard. A starter kit with dataloaders and YOLOv8 baselines are available to get you up and running quickly. Now it’s your turn to beat them.
The CodaBench Robust Thermal Image Object Detection Challenge Platform can be found here: https://www.codabench.org/competitions/10954/
Challenge Schedule:
- October 17th AoE: Start of competition and Development Phase
- December 1st AoE: Start of Testing Phase and end of Development Phase
- December 7th AoE: End of competition
- December 14th AoE: Paper submission deadline (Challenge Participants only)
- December 23rd AoE: Decision notification and announcement of challenge winner
- January 9th 11:59 PM, PT: Camera-ready deadline
This challenge is organized by Anders S. Johansen (Aalborg University, Denmark), Andreas Aakerberg, and Kamal Nasrollahi (Aalborg University, and Milestone Systems, Denmark), together with Marco Parola and Mario G.C.A. Cimino (University of Pisa, Italy).
The challenge is hosted on CodaBench, and opens October 17th, 2025. The LTDv2 dataset used in the challenge can be downloaded here.
