文献类型：期刊文献

英文题名：Attentive feature integration network for detecting salient objects in images

作者：Zhang, Qing[1];Cui, Wenzhao[1];Shi, Yanjiao[1];Zhang, Xueqin[2];Liu, Yunxiang[1]

机构：[1]Shanghai Inst Technol, Dept Comp Sci & Informat Engn, Shanghai, Peoples R China;[2]East China Univ Sci & Technol, Dept Informat Sci & Engn, Shanghai, Peoples R China

年份：2020

卷号：411

起止页码：268

外文期刊名：NEUROCOMPUTING

收录：;EI(收录号：20202708898947);WOS:【SCI-EXPANDED(收录号:WOS:000571895700008)】；

基金：This work is supported by Natural Science Foundation of Shanghai under Grant No. 19ZR1455300, Science and Technology Development Foundation of Shanghai Institute of Technology under Grant No. ZQ2018-23, and National Natural Science Foundation of China under Grant No. 61806126.

语种：英文

外文关键词：Salient object detection; Saliency detection; Channel-wise attention; Deep supervision; Multi-level feature integration

摘要：Benefiting from the development of convolutional neural networks, salient object detection has yielded a qualitative leap in performance. In recent years, most of deep learning based methods utilize multi-level features and obtain inferred saliency map in a coarse-to-fine manner. However, how to learn and represent powerful features is still a challenge. In this paper, we propose a novel FCN-like approach named attentive feature integration network (AFINet) for pixel-wise salient object detection, which results saliency maps with explicit boundary and uniform highlighted regions. Specifically, it adopts feature enhancement module (FEM) to extract rich and enhanced features from backbone net. A feature discrimination module (FDM) is designed to utilize the predicted saliency map generated by deeper layer to help shallower layer learn useful and discriminative attentive features. Moreover, we introduce the saliency information from deeper layer to the shallower one in saliency prediction module (SPM), which helps shallow side outputs accurately locate salient regions. In addition, we design a saliency fusion module (SFM) to integrate different side outputs for utilizing multi-level features. Finally, a fully connected CRF scheme can be optimally incorporated for obtaining saliency results with a higher accuracy. Both qualitative and quantitative comparisons and evaluations conducted on five publicly benchmark datasets demonstrate that our proposed approach compares favorably against 17 state-of-the-art approaches. (c) 2020 Elsevier B.V. All rights reserved.