On zero-shot recognition of generic objects

被引:6
|
作者
Hascoet, Tristan [1 ]
Ariki, Yasuo [1 ]
Takiguchi, Tetsuya [1 ]
机构
[1] Kobe Univ, Grad Sch Syst Informat, Kobe, Hyogo, Japan
关键词
D O I
10.1109/CVPR.2019.00978
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Many recent advances in computer vision are the result of a healthy competition among researchers on high quality, task-specific, benchmarks. After a decade of active research, zero-shot learning (ZSL) models accuracy on the Imagenet benchmark remains far too low to be considered for practical object recognition applications. In this paper, we argue that the main reason behind this apparent lack of progress is the poor quality of this benchmark. We highlight major structural flaws of the current benchmark and analyze different factors impacting the accuracy of ZSL models. We show that the actual classification accuracy of existing ZSL models is significantly higher than was previously thought as we account for these flaws. We then introduce the notion of structural bias specific to ZSL datasets. We discuss how the presence of this new form of bias allows for a trivial solution to the standard benchmark and conclude on the need for a new benchmark. We then detail the semi-automated construction of a new benchmark to address these flaws.
引用
收藏
页码:9545 / 9553
页数:9
相关论文
共 50 条
  • [41] Extreme Reverse Projection Learning for Zero-Shot Recognition
    Guan, Jiechao
    Zhao, An
    Lu, Zhiwu
    COMPUTER VISION - ACCV 2018, PT I, 2019, 11361 : 125 - 141
  • [42] Zero-Shot Grounding of Objects from Natural Language Queries
    Sadhu, Arka
    Chen, Kan
    Nevatia, Ram
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 4693 - 4702
  • [43] Global Semantic Descriptors for Zero-Shot Action Recognition
    Estevam, Valter
    Laroca, Rayson
    Pedrini, Helio
    Menotti, David
    IEEE SIGNAL PROCESSING LETTERS, 2022, 29 : 1843 - 1847
  • [44] Human Motion Recognition Using Zero-Shot Learning
    Mohammadi, Farid Ghareh
    Imteaj, Ahmed
    Amini, M. Hadi
    Arabnia, Hamid R.
    ADVANCES IN ARTIFICIAL INTELLIGENCE AND APPLIED COGNITIVE COMPUTING, 2021, : 171 - 181
  • [45] Variational Autoencoder for Zero-Shot Recognition of Bai Characters
    Lin, Weiwei
    Ma, Tai
    Zhang, Zeqing
    Li, Xiaofan
    Xue, Xingsi
    WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2022, 2022
  • [46] Orthogonal Temporal Interpolation for Zero-Shot Video Recognition
    Zhu, Yan
    Zhuo, Junbao
    Ma, Bin
    Geng, Jiajia
    Wei, Xiaoming
    Wei, Xiaolin
    Wang, Shuhui
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2023, 2023, : 7491 - 7501
  • [47] SEMANTIC EMBEDDING SPACE FOR ZERO-SHOT ACTION RECOGNITION
    Xu, Xun
    Hospedales, Timothy
    Gong, Shaogang
    2015 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2015, : 63 - 67
  • [48] EXPLORING SYNONYMS AS CONTEXT IN ZERO-SHOT ACTION RECOGNITION
    Alexiou, Ioannis
    Xiang, Tao
    Gong, Shaogang
    2016 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2016, : 4190 - 4194
  • [49] Dissimilarity Representation Learning for Generalized Zero-Shot Recognition
    Yang, Gang
    Liu, Jinlu
    Xu, Jieping
    Li, Xirong
    PROCEEDINGS OF THE 2018 ACM MULTIMEDIA CONFERENCE (MM'18), 2018, : 2032 - 2039
  • [50] JOINT PROJECTION AND SUBSPACE LEARNING FOR ZERO-SHOT RECOGNITION
    Liu, Guangzhen
    Guan, Jiechao
    Zhang, Manli
    Zhang, Jianhong
    Wang, Zihao
    Lu, Zhiwu
    2019 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME), 2019, : 1228 - 1233