This paper presents a DETR-based method for cross-domain weakly supervised object detection (CDWSOD), aiming at adapting the detector from source to target domain through weak supervision. We think DETR has strong potential for CDWSOD due to an insight: the encoder and the decoder in DETR are both based on the attention mechanism and are thus capable of aggregating semantics across the entire image. The aggregation results, i.e., image-level predictions, can naturally exploit the weak supervision for domain alignment. Such motivated, we propose DETR with additional Global Aggregation (DETR-GA), a CDWSOD detector that simultaneously makes "instance-level + image-level" predictions and utilizes "strong + weak" supervisions. The key point of DETR-GA is very simple: for the encoder / decoder, we respectively add multiple class queries / a foreground query to aggregate the semantics into image-level predictions. Our query-based aggregation has two advantages. First, in the encoder, the weakly-supervised class queries are capable of roughly locating the corresponding positions and excluding the distraction from non-relevant regions. Second, through our design, the object queries and the foreground query in the decoder share consensus on the class semantics, therefore making the strong and weak supervision mutually benefit each other for domain alignment. Extensive experiments on four popular cross-domain benchmarks show that DETR-GA significantly improves CSWSOD and advances the states of the art (e.g., 29.0% --> 79.4% mAP on PASCAL VOC --> Clipart_all dataset).

翻译：本文提出一种基于DETR的跨域弱监督目标检测方法（CDWSOD），旨在通过弱监督实现检测器从源域到目标域的适应性迁移。我们认为DETR在CDWSOD领域具有巨大潜力，其核心洞察在于：DETR中的编码器与解码器均基于注意力机制，因此具备聚合整幅图像语义的能力。这种聚合结果（即图像级预测）可自然地利用弱监督信号实现域对齐。基于这一动机，我们提出带有全局聚合机制的DETR（DETR-GA）——一种同时进行"实例级+图像级"预测并利用"强监督+弱监督"信号的CDWSOD检测器。DETR-GA的关键设计极其简洁：对编码器/解码器分别添加多类查询/前景查询，将语义聚合为图像级预测。这种基于查询的聚合方法具有两点优势：其一，在编码器中，弱监督分类查询能够粗略定位对应区域并排除无关区域的干扰；其二，通过我们的设计，解码器中的目标查询与前景查询在类别语义上达成共识，使得强监督与弱监督信号在域对齐过程中相互促进。在四个主流跨域基准数据集上的大量实验表明，DETR-GA显著提升了CSWSOD性能并刷新了当前最优水平（例如在PASCAL VOC→Clipart_all数据集上，mAP从29.0%提升至79.4%）。