（论文阅读）Towards Universal Object Detection by Domain Attention

问题：目前的目标检测器只能适用于单一的检测场景。

目的：建立一种适用于各种场景的通用的目标检测系统。

（a）单域检测器组 （b）自适应多域检测器 （c）通用检测器 （d）域注意通用检测器 

相关研究：

目标检测：两阶段检测框架：RCNN,Fast R-CNN,Faster R-CNN；一阶段检测框架（速度快）：YOLO,SSD。它们应用在不同数据集上时都需要微调模型。

多任务学习：多任务学习研究如何在一个输入域内同时学习多个任务。

多域学习/适应：多域学习在已知先验知识时，解决多个域的表示学习问题。

注意模型：提到了SE模块。

建立11个数据集通用的目标检测标准：(1)Pascal VOC:Mark Everingham, SM Ali Eslami, Luc Van Gool, Christo- pher KI Williams, John Winn, and Andrew Zisserman. The pascal visual object classes challenge: A retrospective. Inter- national journal ofcomputer vision, 111(1):98–136, 2015.(2)WiderFace:Shuo Yang, Ping Luo, Chen-Change Loy, and Xiaoou Tang. Wider face: A face detection benchmark. In CVPR, pages 5525–5533, 2016.(3)KITTI:Andreas Geiger, Philip Lenz, and Raquel Urtasun. Are we ready for autonomous driving? the kitti vision benchmark suite. In CVPR, pages 3354–3361, 2012.(4)LISA:Andreas Møgelmose, Mohan M Trivedi, and Thomas B Moeslund. Vision-based traffic sign detection and analy- sis for intelligent driver assistance systems: Perspectives and survey. IEEE Trans. Intelligent Transportation Systems, 13(4):1484–1497, 2012.(5)DOTA:Gui-Song Xia, Xiang Bai, Jian Ding, Zhen Zhu, Serge Be- longie, Jiebo Luo, Mihai Datcu, Marcello Pelillo, and Liang- pei Zhang. Dota: A large-scale dataset for object detection in aerial images. In Proc. CVPR, 2018.(6)COCO:Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll´ar, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In ECCV, pages 740–755, 2014.(7)Watercolor:Naoto Inoue, Ryosuke Furuta, Toshihiko Yamasaki, and Kiy- oharu Aizawa. Cross-domain weakly-supervised object de- tection through progressive domain adaptation. In CVPR, pages 5001–5009, 2018.(8)Clipart:Naoto Inoue, Ryosuke Furuta, Toshihiko Yamasaki, and Kiy- oharu Aizawa. Cross-domain weakly-supervised object de- tection through progressive domain adaptation. In CVPR, pages 5001–5009, 2018.(9)Comic:Naoto Inoue, Ryosuke Furuta, Toshihiko Yamasaki, and Kiy- oharu Aizawa. Cross-domain weakly-supervised object de- tection through progressive domain adaptation. In CVPR, pages 5001–5009, 2018.(10)Kitchen:Georgios Georgakis, Md Alimoor Reza, Arsalan Mousavian, Phi-Hung Le, and Jana Kosecka. Multiview rgb-d dataset for object instance detection. arXiv preprint arXiv:1609.07826, 2016.(11)DeepLesions:Ke Yan, Xiaosong Wang, Le Lu, Ling Zhang, Adam Har- rison, Mohammadhadi Bagheri, and Ronald M Summers. Deep lesion graphs in the wild: relationship learning and organization of significant radiology image findings in a di- verse large-scale lesion database. In IEEE CVPR, 2018.

自适应多域检测器：

前面的参数是共享的，全连接层做成开关型应用于不同的域（但是需要用到感兴趣区域的先验知识）。 

通用检测器（通用SE适配器组+域注意）：

所有的任务共享一个检测器， 输出层会针对不同的域。也就是说除了输出层，所有的参数都是共享的。

补充（SE）：