A Weakly Supervised Learning Framework for Salient Object Detection via Hybrid Labels

作     者：Cong, Runming Qin, Qi Zhang, Chen Jiang, Qiuping Wang, Shiqi Zhao, Yao Kwong, Sam 

作者机构：Institute of Information Science Beijing Jiaotong University Beijing100044 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China Department of Computer Science City University of Hong Kong Hong Kong School of Information Science and Engineering Ningbo University Ningbo315211 China City University of Hong Kong Shenzhen Research Institute Shenzhen51800 China 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Benchmarking 

摘      要：Fully-supervised salient object detection (SOD) methods have made great progress, but such methods often rely on a large number of pixel-level annotations, which are time-consuming and labour-intensive. In this paper, we focus on a new weakly-supervised SOD task under hybrid labels, where the supervision labels include a large number of coarse labels generated by the traditional unsupervised method and a small number of real labels. To address the issues of label noise and quantity imbalance in this task, we design a new pipeline framework with three sophisticated training strategies. In terms of model framework, we decouple the task into label refinement sub-task and salient object detection sub-task, which cooperate with each other and train alternately. Specifically, the R-Net is designed as a two-stream encoder-decoder model equipped with Blender with Guidance and Aggregation Mechanisms (BGA), aiming to rectify the coarse labels for more reliable pseudo-labels, while the S-Net is a replaceable SOD network supervised by the pseudo labels generated by the current R-Net. Note that, we only need to use the trained S-Net for testing. Moreover, in order to guarantee the effectiveness and efficiency of network training, we design three training strategies, including alternate iteration mechanism, group-wise incremental mechanism, and credibility verification mechanism. Experiments on five SOD benchmarks show that our method achieves competitive performance against weakly-supervised/unsupervised methods both qualitatively and quantitatively. The code and results can be found from the link of https://***/proj_***. © 2022, CC BY-NC-SA.