D3T: Distinctive Dual-Domain Teacher Zigzagging Across RGB-Thermal Gap for Domain-Adaptive Object Detection
This addresses the problem of adapting object detection models to thermal imagery for applications like surveillance or autonomous driving, representing an incremental advance in a niche area.
The paper tackles domain adaptation for object detection from visible to thermal domains, which has a large gap, by proposing a Distinctive Dual-Domain Teacher framework with zigzag learning, achieving improved performance on FLIR and KAIST datasets.
Domain adaptation for object detection typically entails transferring knowledge from one visible domain to another visible domain. However, there are limited studies on adapting from the visible to the thermal domain, because the domain gap between the visible and thermal domains is much larger than expected, and traditional domain adaptation can not successfully facilitate learning in this situation. To overcome this challenge, we propose a Distinctive Dual-Domain Teacher (D3T) framework that employs distinct training paradigms for each domain. Specifically, we segregate the source and target training sets for building dual-teachers and successively deploy exponential moving average to the student model to individual teachers of each domain. The framework further incorporates a zigzag learning method between dual teachers, facilitating a gradual transition from the visible to thermal domains during training. We validate the superiority of our method through newly designed experimental protocols with well-known thermal datasets, i.e., FLIR and KAIST. Source code is available at https://github.com/EdwardDo69/D3T .