![[2602.22621] CGSA: Class-Guided Slot-Aware Adaptation for Source-Free Object Detection](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.22621] CGSA: Class-Guided Slot-Aware Adaptation for Source-Free Object Detection
Summary
The paper presents CGSA, a novel framework for Source-Free Domain Adaptive Object Detection that integrates object-centric learning to enhance domain adaptation performance.
Why It Matters
As object detection continues to evolve, the ability to adapt models to new domains without retaining source data is crucial, especially in privacy-sensitive applications. CGSA introduces innovative techniques that could significantly improve detection accuracy in such scenarios, making it relevant for researchers and practitioners in the field.
Key Takeaways
- CGSA integrates Hierarchical Slot Awareness for improved object representation.
- The Class-Guided Slot Contrast module enhances semantic consistency.
- Extensive experiments show CGSA outperforms existing SF-DAOD methods.
- The framework is particularly useful in privacy-sensitive adaptation scenarios.
- The code for CGSA is publicly available for further research and application.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.22621 (cs) [Submitted on 26 Feb 2026] Title:CGSA: Class-Guided Slot-Aware Adaptation for Source-Free Object Detection Authors:Boyang Dai, Zeng Fan, Zihao Qi, Meng Lou, Yizhou Yu View a PDF of the paper titled CGSA: Class-Guided Slot-Aware Adaptation for Source-Free Object Detection, by Boyang Dai and 4 other authors View PDF HTML (experimental) Abstract:Source-Free Domain Adaptive Object Detection (SF-DAOD) aims to adapt a detector trained on a labeled source domain to an unlabeled target domain without retaining any source data. Despite recent progress, most popular approaches focus on tuning pseudo-label thresholds or refining the teacher-student framework, while overlooking object-level structural cues within cross-domain data. In this work, we present CGSA, the first framework that brings Object-Centric Learning (OCL) into SF-DAOD by integrating slot-aware adaptation into the DETR-based detector. Specifically, our approach integrates a Hierarchical Slot Awareness (HSA) module into the detector to progressively disentangle images into slot representations that act as visual priors. These slots are then guided toward class semantics via a Class-Guided Slot Contrast (CGSC) module, maintaining semantic consistency and prompting domain-invariant adaptation. Extensive experiments on multiple cross-domain datasets demonstrate that our approach outperforms previous SF-DAOD methods, with theoretical derivatio...