This is the official Repository of the RTIOD - Robust Thermal-Image Object Detection Challenge at WACV 2026 (RTIOD@WACV2026)

Ready to stress-test your detector in the wild? The 6th Real World Surveillance Workshop at WACV 2026 RWS@WACV2026 introduces a new challenge on Robust Thermal-Image Object Detection using the LTDv2 dataset 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, which slowly erode detector performance.

The objectives, rules, and challenge submission are handled on the Codabench challenge page.

Challenge dataset: LTDv2 paper dataset

A starting kit to run some baseline experiments with YOLO is available here, but feel free to reimplement from scratch all the code ;) In that case, we also release the torch dataset to manage the dataset in the original COCO format along with some configuration, and utility functions.

Associated workshop at WACV2026: Real-World Surveillance: Applications and Challenges Workshop, 6th

• October 17th: Start of competition and Development Phase
• December 1st: Start of Testing Phase and end of Development Phase
• December 7th: End of competition
• December 7th: Paper submission deadline
• December 14th: Paper submission deadline (Challenge Participants)
• December 23rd: Decision notification
• January 9th 11:59 PM, PT: Camera-ready deadline