Semantic-aligned Fusion Transformer for One-shot Object Detection
About
One-shot object detection aims at detecting novel objects according to merely one given instance. With extreme data scarcity, current approaches explore various feature fusions to obtain directly transferable meta-knowledge. Yet, their performances are often unsatisfactory. In this paper, we attribute this to inappropriate correlation methods that misalign query-support semantics by overlooking spatial structures and scale variances. Upon analysis, we leverage the attention mechanism and propose a simple but effective architecture named Semantic-aligned Fusion Transformer (SaFT) to resolve these issues. Specifically, we equip SaFT with a vertical fusion module (VFM) for cross-scale semantic enhancement and a horizontal fusion module (HFM) for cross-sample feature fusion. Together, they broaden the vision for each feature point from the support to a whole augmented feature pyramid from the query, facilitating semantic-aligned associations. Extensive experiments on multiple benchmarks demonstrate the superiority of our framework. Without fine-tuning on novel classes, it brings significant performance gains to one-stage baselines, lifting state-of-the-art results to a higher level.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Object Detection | MS-COCO 2017 (val) | Base Avg AP5048.3 | 27 | |
| Object Detection | COCO 2017 (Split-3) | Base AP5047.9 | 6 | |
| Object Detection | COCO 2017 (Split-4) | bAP5049 | 6 | |
| Object Detection | COCO Average across splits 2017 (Avg) | bAP5048.3 | 6 | |
| Object Detection | COCO 2017 (Split-1) | Base AP5049.2 | 6 | |
| Object Detection | COCO 2017 (Split-2) | bAP5047.2 | 6 | |
| Object Detection | PASCAL VOC Base classes 2007 (test) | AP (Plant)59.7 | 5 | |
| Object Detection | PASCAL VOC Novel classes 2007 (test) | Cow88.1 | 5 |