papers AI Learner
The Github is limit! Click to go to the new site.

Hierarchical Structure and Joint Training for Large Scale Semi-supervised Object Detection

2019-05-30
Ye Guo, Yali Li, Shengjin Wang

Abstract

Generic object detection is one of the most fundamental and important problems in computer vision. When it comes to large scale object detection for thousands of categories, it is unpractical to provide all the bounding box labels for each category. In this paper, we propose a novel hierarchical structure and joint training framework for large scale semi-supervised object detection. First, we utilize the relationships among target categories to model a hierarchical network to further improve the performance of recognition. Second, we combine bounding-box-level labeled images and image-level labeled images together for joint training, and the proposed method can be easily applied in current two-stage object detection framework with excellent performance. Experimental results show that the proposed large scale semi-supervised object detection network obtains the state-of-the-art performance, with the mAP of 38.1% on the ImageNet detection validation dataset.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1905.12863

PDF

http://arxiv.org/pdf/1905.12863


Similar Posts

Comments