Efficient Anomaly Detection Using Self-Supervised Multi-Cue Tasks

被引:11
|
作者
Jezequel, Loic [1 ,2 ]
Vu, Ngoc-Son [1 ]
Beaudet, Jean [2 ]
Histace, Aymeric [1 ]
机构
[1] CY Cergy Paris Univ, ENSEA, CNRS, ETIS UMR 8051, F-95000 Paris, France
[2] Idemia Ident & Secur, F-95520 Osny, France
关键词
Task analysis; Anomaly detection; Training; Feature extraction; Self-supervised learning; Faces; Neural networks; fine grained classification; self-supervised learning; multi-task learning; one-class learning;
D O I
10.1109/TIP.2022.3231532
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Anomaly detection is important in many real-life applications. Recently, self-supervised learning has greatly helped deep anomaly detection by recognizing several geometric transformations. However these methods lack finer features, usually highly depend on the anomaly type, and do not perform well on fine-grained problems. To address these issues, we first introduce in this work three novel and efficient discriminative and generative tasks which have complementary strength: (i) a piece-wise jigsaw puzzle task focuses on structure cues; (ii) a tint rotation recognition is used within each piece, taking into account the colorimetry information; (iii) and a partial re-colorization task considers the image texture. In order to make the re-colorization task more object-oriented than background-oriented, we propose to include the contextual color information of the image border via an attention mechanism. We then present a new out-of-distribution detection function and highlight its better stability compared to existing methods. Along with it, we also experiment different score fusion functions. Finally, we evaluate our method on an extensive protocol composed of various anomaly types, from object anomalies, style anomalies with fine-grained classification to local anomalies with face anti-spoofing datasets. Our model significantly outperforms state-of-the-art with up to 36% relative error improvement on object anomalies and 40% on face anti-spoofing problems.
引用
收藏
页码:807 / 821
页数:15
相关论文
共 50 条
  • [21] Self-Supervised Masking for Unsupervised Anomaly Detection and Localization
    Huang, Chaoqin
    Xu, Qinwei
    Wang, Yanfeng
    Wang, Yu
    Zhang, Ya
    IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 4426 - 4438
  • [22] Self-supervised Sparse Representation for Video Anomaly Detection
    Wu, Jhih-Ciang
    Hsieh, He-Yen
    Chen, Ding-Jie
    Fuh, Chiou-Shann
    Liu, Tyng-Luh
    COMPUTER VISION, ECCV 2022, PT XIII, 2022, 13673 : 729 - 745
  • [23] Self-supervised enhanced denoising diffusion for anomaly detection
    Li, Shu
    Yu, Jiong
    Lu, Yi
    Yang, Guangqi
    Du, Xusheng
    Liu, Su
    INFORMATION SCIENCES, 2024, 669
  • [24] CutPaste: Self-Supervised Learning for Anomaly Detection and Localization
    Li, Chun-Liang
    Sohn, Kihyuk
    Yoon, Jinsung
    Pfister, Tomas
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 9659 - 9669
  • [25] Multi-Cue Onboard Pedestrian Detection
    Wojek, Christian
    Walk, Stefan
    Schiele, Bernt
    CVPR: 2009 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOLS 1-4, 2009, : 794 - 801
  • [26] A Novel Multi-cue Integration System for Efficient Human Fall Detection
    Wang, Xue
    Liu, Hong
    Liu, Mengyuan
    2016 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO), 2016, : 1319 - 1324
  • [27] Road Condition Anomaly Detection using Self-Supervised Learning from Audio
    Gim, U-Ju
    2023 IEEE 26TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, ITSC, 2023, : 675 - 680
  • [28] A Self-Supervised Learning Approach to Road Anomaly Detection Using Masked Autoencoders
    Dutta, Proma
    Podder, Kanchon Kanti
    Zhang, Jian
    Hecht, Christian
    Swarna, Surya
    Bhavsar, Parth
    INTERNATIONAL CONFERENCE ON TRANSPORTATION AND DEVELOPMENT 2024: PAVEMENTS AND INFRASTRUCTURE SYSTEMS, ICTD 2024, 2024, : 536 - 547
  • [29] Self-Supervised Learning in Medical Imaging: Anomaly Detection in MRI Using Autoencoders
    Wismueller, Axel
    Vosoughi, M. Ali
    REAL-TIME IMAGE PROCESSING AND DEEP LEARNING 2022, 2022, 12102
  • [30] Anomaly Detection on the Rail Lines Using Semantic Segmentation and Self-supervised Learning
    Jahan, Kanwal
    Umesh, Jeethesh Pai
    Roth, Michael
    2021 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2021), 2021,