OAFormer: Occlusion Aware Transformer for Camouflaged Object Detection

被引:12
作者
Yang, Xin [1 ]
Zhu, Hengliang [1 ]
Mao, Guojun [1 ]
Xing, Shuli [1 ]
机构
[1] FuJian Univ Technol, Coll Comp Sci & Math, Fuzhou, Peoples R China
来源
2023 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, ICME | 2023年
基金
中国国家自然科学基金;
关键词
camouflaged object detection; hierarchical location guidance; transformer; occlusion;
D O I
10.1109/ICME55011.2023.00246
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Camouflaged object usually have a similar appearance or color to their surrounding environment, so it's difficult to be detected, especially in heavily obscured situations. To deal with this challenge, in this paper, we propose a novel occlusion aware transformer network (OAFormer) to accurately identify the occluded camouflaged object. In OAFormer, a hierarchical location guidance module (HLGM) is designed to locate the potential locations of camouflaged objects. Then, in order to perceive the structural consistency of the occluded object, we design a neighborhood searching module (NSM) to focus on local pixel details of concealed objects. Besides, for each NSM, we take advantages of transformer blocks to capture long-distance dependencies. So our model can easily capture the complete camouflaged object. In the end, we utilize the auxiliary supervision strategy to promote the learning ability of our model. Compared with other state-of-the-art methods, the proposed OAFormer achieves higher accuracy on four challenging datasets. Code and models are available at: https://github.com/xinyang920/OAFormer.
引用
收藏
页码:1421 / 1426
页数:6
相关论文
共 33 条
[1]  
Dai S, 2017, INT CONF ACOUST SPEE, P1028, DOI 10.1109/ICASSP.2017.7952312
[2]  
Deng J, 2009, PROC CVPR IEEE, P248, DOI 10.1109/CVPRW.2009.5206848
[3]  
Deng-Ping Fan, 2020, Medical Image Computing and Computer Assisted Intervention - MICCAI 2020. 23rd International Conference. Proceedings. Lecture Notes in Computer Science (LNCS 12266), P263, DOI 10.1007/978-3-030-59725-2_26
[4]  
Fan D.-P., 2021, Scientia Sinica Informationis, V6
[5]   Concealed Object Detection [J].
Fan, Deng-Ping ;
Ji, Ge-Peng ;
Cheng, Ming-Ming ;
Shao, Ling .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (10) :6024-6042
[6]   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
[7]   Camouflaged Object Detection [J].
Fan, Deng-Ping ;
Ji, Ge-Peng ;
Sun, Guolei ;
Cheng, Ming-Ming ;
Shen, Jianbing ;
Shao, Ling .
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, :2774-2784
[8]  
Huerta I, 2007, LECT NOTES COMPUT SC, V4478, P475
[9]  
Islam MA., 2020, ARXIV PREPRINT ARXIV
[10]   UACANet: Uncertainty Augmented Context Attention for Polyp Segmentation [J].
Kim, Taehun ;
Lee, Hyemin ;
Kim, Daijin .
PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2021, 2021, :2167-2175