Meta-DETR: Image-Level Few-Shot Detection with Inter-Class Correlation Exploitation
About
Few-shot object detection has been extensively investigated by incorporating meta-learning into region-based detection frameworks. Despite its success, the said paradigm is still constrained by several factors, such as (i) low-quality region proposals for novel classes and (ii) negligence of the inter-class correlation among different classes. Such limitations hinder the generalization of base-class knowledge for the detection of novel-class objects. In this work, we design Meta-DETR, which (i) is the first image-level few-shot detector, and (ii) introduces a novel inter-class correlational meta-learning strategy to capture and leverage the correlation among different classes for robust and accurate few-shot object detection. Meta-DETR works entirely at image level without any region proposals, which circumvents the constraint of inaccurate proposals in prevalent few-shot detection frameworks. In addition, the introduced correlational meta-learning enables Meta-DETR to simultaneously attend to multiple support classes within a single feedforward, which allows to capture the inter-class correlation among different classes, thus significantly reducing the misclassification over similar classes and enhancing knowledge generalization to novel classes. Experiments over multiple few-shot object detection benchmarks show that the proposed Meta-DETR outperforms state-of-the-art methods by large margins. The implementation codes are available at https://github.com/ZhangGongjie/Meta-DETR.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Object Detection | PASCAL VOC Novel Set 3 2007+2012 | mAP5060.6 | 139 | |
| Object Detection | PASCAL VOC 2007+2012 (Novel Set 1) | -- | 75 | |
| Object Detection | PASCAL VOC Novel Set 2 2007+2012 | -- | 75 | |
| Few-shot Object Detection | MS COCO novel classes | mAP22.2 | 37 | |
| Object Detection | COCO-UniFS (val) | AP15.3 | 24 | |
| Instance Segmentation | COCO-UniFS (val) | AP8.1 | 14 |