Texture-guided Saliency Distilling for Unsupervised Salient Object Detection

被引:28
作者
Zhou, Huajun [1 ]
Qiao, Bo [1 ]
Yang, Lingxiao [1 ]
Lai, Jianhuang [1 ,2 ,3 ]
Xie, Xiaohua [1 ,2 ,3 ]
机构
[1] Sun Yat Sen Univ, Sch Comp Sci & Engn, Guangzhou, Peoples R China
[2] Guangdong Prov Key Lab Informat Secur Technol, Shenzhen, Peoples R China
[3] Minist Educ, Key Lab Machine Intelligence & Adv Comp, Guangzhou, Peoples R China
来源
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR | 2023年
基金
中国国家自然科学基金;
关键词
RECOGNITION;
D O I
10.1109/CVPR52729.2023.00701
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Deep Learning-based Unsupervised Salient Object Detection (USOD) mainly relies on the noisy saliency pseudo labels that have been generated from traditional handcraft methods or pre-trained networks. To cope with the noisy labels problem, a class of methods focus on only easy samples with reliable labels but ignore valuable knowledge in hard samples. In this paper, we propose a novel USOD method to mine rich and accurate saliency knowledge from both easy and hard samples. First, we propose a Confidence-aware Saliency Distilling (CSD) strategy that scores samples conditioned on samples' confidences, which guides the model to distill saliency knowledge from easy samples to hard samples progressively. Second, we propose a Boundary-aware Texture Matching (BTM) strategy to refine the boundaries of noisy labels by matching the textures around the predicted boundaries. Extensive experiments on RGB, RGBD, RGB-T, and video SOD benchmarks prove that our method achieves state-of-the-art USOD performance. Code is available at www.github.com/moothes/A2S-v2.
引用
收藏
页码:7257 / 7267
页数:11
相关论文
共 75 条
  • [21] Deeply Supervised Salient Object Detection with Short Connections
    Hou, Qibin
    Cheng, Ming-Ming
    Hu, Xiaowei
    Borji, Ali
    Tu, Zhuowen
    Torr, Philip H. S.
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2019, 41 (04) : 815 - 828
  • [22] Hu J, 2018, PROC CVPR IEEE, P7132, DOI [10.1109/TPAMI.2019.2913372, 10.1109/CVPR.2018.00745]
  • [23] Ji W., 2022, arXiv preprint arXiv:2205.07179
  • [24] Saliency Detection via Absorbing Markov Chain
    Jiang, Bowen
    Zhang, Lihe
    Lu, Huchuan
    Yang, Chuan
    Yang, Ming-Hsuan
    [J]. 2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2013, : 1665 - 1672
  • [25] Ju R, 2014, IEEE IMAGE PROC, P1115, DOI 10.1109/ICIP.2014.7025222
  • [26] Video Segmentation by Tracking Many Figure-Ground Segments
    Li, Fuxin
    Kim, Taeyoung
    Humayun, Ahmad
    Tsai, David
    Rehg, James M.
    [J]. 2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2013, : 2192 - 2199
  • [27] Li GB, 2015, PROC CVPR IEEE, P5455, DOI 10.1109/CVPR.2015.7299184
  • [28] The Secrets of Salient Object Segmentation
    Li, Yin
    Hou, Xiaodi
    Koch, Christof
    Rehg, James M.
    Yuille, Alan L.
    [J]. 2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2014, : 280 - 287
  • [29] Liao Guibiao, 2022, IEEE T CIRCUITS SYST
  • [30] Lin Xiangru, 2022, 36 AAAI C ART INT