1. Collaborative learning in bounding box regression for object detection
- Author
-
Hongpeng Wang, Xiuli Shao, Xian Fang, Ruixun Zhang, and Zengsheng Kuang
- Subjects
Computer science ,Intersection (set theory) ,business.industry ,Pattern recognition ,Collaborative learning ,02 engineering and technology ,01 natural sciences ,Object detection ,Artificial Intelligence ,Minimum bounding box ,Bounding overwatch ,Component (UML) ,0103 physical sciences ,Signal Processing ,Metric (mathematics) ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Computer Vision and Pattern Recognition ,Artificial intelligence ,010306 general physics ,Representation (mathematics) ,business ,Software - Abstract
Object detection has attracted growing attention due to its extensive application prospect, in which bounding box regression is an essential component. Dedicated to collaborative learning in bounding box regression, we explore the unified framework of smooth l 1 and intersection over union, named SLIoU. On the basis of that, we propose a SLIoU loss as localization loss, which focuses on the geometric relationships of pairs of rectangular bounding boxes in overlapping degree, central position and structural shape. Furthermore, we propose a SLIoU-NMS for suppressing redundant detection boxes, which adaptively maps the evaluation value of detection boxes to meet the evaluation metric using nonlinear representation. By incorporating SLIoU loss and SLIoU-NMS into the state-of-the-art one-stage detectors, the detection performance is considerably improved.
- Published
- 2021