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.
翻译:通过将元学习纳入区域探测框架,对很少见的物体探测进行了广泛调查。尽管取得了成功,但上述模式仍然受到若干因素的制约,例如(一) 低质量区域新类建议和(二) 不同类别之间不同类别之间关联的疏忽,这些限制妨碍了基础级知识的普及,以探测新类物体。在这项工作中,我们设计了Meta-DETR, (一) 是第一个图像级小片检测器,和(二) 引入了一个新的等级间相关元学习战略,以捕捉和利用不同类别之间的相关性,以进行强力和准确的少发物体探测。Meta-DETR完全在图像一级开展工作,而没有任何区域建议,从而避免了流行的少发检测框架中不准确建议之间的制约。此外,引入了相关性的元数据学习,使Met-DETR能够同时参加一个反馈前的多个支助类别,从而能够捕捉到不同类别之间的不同类别间关联,从而大大减少类似类别之间的误分类,并将知识推广到新类别。Met-DETR完全在图像/Mialfrodal-comstational-commatial 正在对多个数据库进行实验。