Deep Sub-Region Network for Salient Object Detection

被引:47
作者
Wang, Liansheng [1 ]
Chen, Rongzhen [1 ]
Zhu, Lei [2 ]
Xie, Haoran [3 ]
Li, Xiaomeng [2 ]
机构
[1] Xiamen Univ, Sch Informat, Dept Comp Sci, Xiamen 361005, Peoples R China
[2] Chinese Univ Hong Kong, Dept Comp Sci & Engn, Hong Kong, Peoples R China
[3] Lingnan Univ, Dept Comp & Decis Sci, Hong Kong, Peoples R China
基金
中国国家自然科学基金;
关键词
Feature extraction; Saliency detection; Object detection; Semantics; Benchmark testing; Visualization; Convolutional neural networks; deep subregion learning; region dilated blocks; parallel atrous spatial pyramid pooling (ASPP) modules; FRAMEWORK; IMAGE; MODEL;
D O I
10.1109/TCSVT.2020.2988768
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Saliency detection is a fundamental and challenging task in computer vision, which aims at distinguishing the most conspicuous objects or regions in an image. Existing deep-learning methods mainly rely on the entire image to learn the global context information for saliency detection, which loses the spatial relation and results in ambiguity in predicting saliency maps. In this paper, we propose a novel deep sub-region network (DSR-Net) equipped with a sequence of sub-region dilated blocks (SRDB) by aggregating multi-scale salient context information of multiple sub-regions, such that the global context information from the whole image and local contexts from sub-regions are fused together, making the saliency prediction more accurate. Our SRDB separates the input feature map at different layers of a convolutional neural network (CNN) into different sub-regions and then designs a parallel ASPP module to refine feature maps at each sub-region. Experiments on the five widely-used saliency benchmark datasets demonstrate that our network outperforms recent state-of-the-art saliency detectors quantitatively and qualitatively on all the benchmarks.
引用
收藏
页码:728 / 741
页数:14
相关论文
共 71 条
[1]  
Achanta R, 2009, PROC CVPR IEEE, P1597, DOI 10.1109/CVPRW.2009.5206596
[2]  
Borji A, 2012, PROC CVPR IEEE, P478, DOI 10.1109/CVPR.2012.6247711
[3]   DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs [J].
Chen, Liang-Chieh ;
Papandreou, George ;
Kokkinos, Iasonas ;
Murphy, Kevin ;
Yuille, Alan L. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2018, 40 (04) :834-848
[4]   Reverse Attention for Salient Object Detection [J].
Chen, Shuhan ;
Tan, Xiuli ;
Wang, Ben ;
Hu, Xuelong .
COMPUTER VISION - ECCV 2018, PT IX, 2018, 11213 :236-252
[5]   Global Contrast based Salient Region Detection [J].
Cheng, Ming-Ming ;
Zhang, Guo-Xin ;
Mitra, Niloy J. ;
Huang, Xiaolei ;
Hu, Shi-Min .
2011 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2011, :409-416
[6]   HSCS: Hierarchical Sparsity Based Co-saliency Detection for RGBD Images [J].
Cong, Runmin ;
Lei, Jianjun ;
Fu, Huazhu ;
Huang, Qingming ;
Cao, Xiaochun ;
Ling, Nam .
IEEE TRANSACTIONS ON MULTIMEDIA, 2019, 21 (07) :1660-1671
[7]   Video Saliency Detection via Sparsity-Based Reconstruction and Propagation [J].
Cong, Runmin ;
Lei, Jianjun ;
Fu, Huazhu ;
Porikli, Fatih ;
Huang, Qingming ;
Hou, Chunping .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2019, 28 (10) :4819-4831
[8]   An Iterative Co-Saliency Framework for RGBD Images [J].
Cong, Runmin ;
Lei, Jianjun ;
Fu, Huazhu ;
Lin, Weisi ;
Huang, Qingming ;
Cao, Xiaochun ;
Hou, Chunping .
IEEE TRANSACTIONS ON CYBERNETICS, 2019, 49 (01) :233-246
[9]  
Deng ZJ, 2018, PROCEEDINGS OF THE TWENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P684
[10]   Structure-measure: A New Way to Evaluate Foreground Maps [J].
Fan, Deng-Ping ;
Cheng, Ming-Ming ;
Liu, Yun ;
Li, Tao ;
Borji, Ali .
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, :4558-4567