ABRA: Teleporting Fine-Tuned Knowledge Across Domains for Open-Vocabulary Object Detection
This addresses the issue of domain adaptation for object detection in practical scenarios lacking annotated data, offering a novel solution but with incremental improvements in a specific domain.
The paper tackles the problem of performance degradation in Open-Vocabulary Object Detection under domain shifts, such as nighttime or foggy scenes, by introducing ABRA, a method that transfers class-specific detection knowledge from a labeled source domain to an unlabeled target domain, achieving successful adaptation across challenging conditions.
Although recent Open-Vocabulary Object Detection architectures, such as Grounding DINO, demonstrate strong zero-shot capabilities, their performance degrades significantly under domain shifts. Moreover, many domains of practical interest, such as nighttime or foggy scenes, lack large annotated datasets, preventing direct fine-tuning. In this paper, we introduce Aligned Basis Relocation for Adaptation(ABRA), a method that transfers class-specific detection knowledge from a labeled source domain to a target domain where no training images containing these classes are accessible. ABRA formulates this adaptation as a geometric transport problem in the weight space of a pretrained detector, aligning source and target domain experts to transport class-specific knowledge. Extensive experiments across challenging domain shifts demonstrate that ABRA successfully teleports class-level specialization under multiple adverse conditions. Our code will be made public upon acceptance.