DFS: A Diverse Feature Synthesis Model for Generalized Zero-Shot Learning

被引:0
|
作者
Li, Bonan [1 ]
Hu, Yinhan [1 ]
Han, Congying [1 ]
Guo, Tiande [1 ]
机构
[1] Univ Chinese Acad Sci, Sch Math Sci, Beijing, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
10.1109/ICPR56361.2022.9956235
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Generative based strategy has shown great potential in the Generalized Zero-Shot Learning task. However, it suffers severe generalization problem due to lacking of feature diversity for unseen classes to train a good classifier. In this paper, we propose to enhance the generalizability of GZSL models via improving feature diversity of unseen classes. For this purpose, we present a novel Diverse Feature Synthesis (DFS) model. Different from prior works that solely utilize semantic knowledge in the generation process, DFS leverages visual knowledge with semantic one in a unified way, thus deriving class-specific diverse feature samples and leading to robust classifier for recognizing both seen and unseen classes in the testing phase. To simplify the learning, DFS represents visual and semantic knowledge in the aligned space, making it able to produce good feature samples with a low-complexity implementation. Accordingly, DFS is composed of two consecutive generators: an aligned feature generator, transferring semantic and visual representations into aligned features; a synthesized feature generator, producing diverse feature samples of unseen classes in the aligned space. We conduct comprehensive experiments to verify the efficacy of DFS. Results demonstrate its effectiveness to generate diverse features for unseen classes, leading to superior performance on multiple benchmarks. Code will be released upon acceptance.
引用
收藏
页码:1851 / 1857
页数:7
相关论文
共 50 条
  • [1] Semantic Feature Extraction for Generalized Zero-Shot Learning
    Kim, Junhan
    Shim, Kyuhong
    Shim, Byonghyo
    [J]. THIRTY-SIXTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FOURTH CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE / THE TWELVETH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2022, : 1166 - 1173
  • [2] FREE: Feature Refinement for Generalized Zero-Shot Learning
    Chen, Shiming
    Wang, Wenjie
    Xia, Beihao
    Peng, Qinmu
    You, Xinge
    Zheng, Feng
    Shao, Ling
    [J]. 2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 122 - 131
  • [3] Unbiased feature generating for generalized zero-shot learning
    Niu, Chang
    Shang, Junyuan
    Huang, Junchu
    Yang, Junmei
    Song, Yuting
    Zhou, Zhiheng
    Zhou, Guoxu
    [J]. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2022, 89
  • [4] Model Selection for Generalized Zero-Shot Learning
    Zhang, Hongguang
    Koniusz, Piotr
    [J]. COMPUTER VISION - ECCV 2018 WORKSHOPS, PT II, 2019, 11130 : 198 - 204
  • [5] Contrastive visual feature filtering for generalized zero-shot learning
    Meng, Shixuan
    Jiang, Rongxin
    Tian, Xiang
    Zhou, Fan
    Chen, Yaowu
    Liu, Junjie
    Shen, Chen
    [J]. INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2024,
  • [6] Towards Discriminative Feature Generation for Generalized Zero-Shot Learning
    Ge, Jiannan
    Xie, Hongtao
    Li, Pandeng
    Xie, Lingxi
    Min, Shaobo
    Zhang, Yongdong
    [J]. IEEE Transactions on Multimedia, 2024, 26 : 10514 - 10529
  • [7] Inference guided feature generation for generalized zero-shot learning
    Han, Zongyan
    Fu, Zhenyong
    Li, Guangyu
    Yang, Jian
    [J]. NEUROCOMPUTING, 2021, 430 : 150 - 158
  • [8] Aligning enhanced feature representation for generalized zero-shot learning
    Zhiyu FANG
    Xiaobin ZHU
    Chun YANG
    Hongyang ZHOU
    Jingyan QIN
    Xu-Cheng YIN
    [J]. Science China(Information Sciences)., 2025, 68 (02) - 88
  • [9] Learning discriminative and representative feature with cascade GAN for generalized zero-shot learning
    Liu, Jingren
    Fu, Liyong
    Zhang, Haofeng
    Ye, Qiaolin
    Yang, Wankou
    Liu, Li
    [J]. KNOWLEDGE-BASED SYSTEMS, 2022, 236
  • [10] Learning discriminative and representative feature with cascade GAN for generalized zero-shot learning
    Liu, Jingren
    Fu, Liyong
    Zhang, Haofeng
    Ye, Qiaolin
    Yang, Wankou
    Liu, Li
    [J]. Knowledge-Based Systems, 2022, 236