Weakly-supervised object detection via mining pseudo ground truth bounding-boxes

PATTERN RECOGNITION

卷:84

Article

Zhang, Yongqiang^1,2  Bai, Yaicheng²  Ding, Mingli¹  Li, Yongqiang¹  Ghanem, Bernard² 

[1] Harbin Inst Technol, Sch Elect Engn & Automat, Harbin 15001, Heilongjiang, Peoples R China

[2] King Abdullah Univ Sci & Technol, Visual Comp Ctr, Thuwal 239556900, Saudi Arabia

关键词: Weakly-supervised learning;    Object detection;    Pseudo ground truth;    Iterative learning;    Deep learning;   

DOI  :  10.1016/j.patcog.2018.07.005

来源: Elsevier

【 摘 要 】

Recently, weakly-supervised object detection has attracted much attention, since it does not require expensive bounding-box annotations while training the network. Although significant progress has also been made, there is still a large gap on the performance between weakly-supervised and fully-supervised object detection. To mitigate this gap, some works try to use the pseudo ground truths generated by a weakly-supervised detector to train a supervised detector. However, such approaches incline to find the most representative parts instead of the whole body of an object, and only seek one ground truth bounding-box per class even though many same-class instances exist in an image. To address these issues, we propose a weakly-supervised to fully-supervised framework (W2F), where a weakly-supervised detector is implemented using multiple instance learning. And then, we propose a pseudo ground-truth excavation (PGE) algorithm to find the accurate pseudo ground truth bounding-box for each instance. Moreover, the pseudo ground-truth adaptation (PGA) algorithm is designed to further refine those pseudo ground truths mined by PGE algorithm. Finally, the mined pseudo ground truths are used as supervision to train a fully-supervised detector. Additionally, we also propose an iterative ground-truth learning (IGL) approach, which enhances the quality of the pseudo ground truths by using the predictions of the fully-supervised detector iteratively. Extensive experiments on the challenging PASCAL VOC 2007 and 2012 benchmarks strongly demonstrate the effectiveness of our method. We obtain 53.1% and 49.4% mAP on VOC2007 and VOC2012 respectively, which is a significant improvement over previous state-of-the-art methods. (C) 2018 Elsevier Ltd. All rights reserved.

【 授权许可】

Free