A Self-supervised GAN for Unsupervised Few-shot Object Recognition

被引:1
|
作者
Nguyen, Khoi [1 ]
Todorovic, Sinisa [1 ]
机构
[1] Oregon State Univ, Corvallis, OR 97330 USA
关键词
D O I
10.1109/ICPR48806.2021.9412539
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper addresses unsupervised few-shot object recognition, where all training images are unlabeled, and test images are divided into queries and a few labeled support images per object class of interest. The training and test images do not share object classes. We extend the vanilla GAN with two loss functions, both aimed at self-supervised learning. The first is a reconstruction loss that enforces the discriminator to reconstruct the probabilistically sampled latent code which has been used for generating the "fake" image. The second is a triplet loss that enforces the discriminator to output image encodings that are closer for more similar images. Evaluation, comparisons, and detailed ablation studies are done in the context of few-shot classification. Our approach significantly outperforms the state of the art on the Mini-Imagenet and Tiered-Imagenet datasets.
引用
收藏
页码:3225 / 3231
页数:7
相关论文
共 50 条
  • [1] A Survey of Self-Supervised and Few-Shot Object Detection
    Huang, Gabriel
    Laradji, Issam
    Vazquez, David
    Lacoste-Julien, Simon
    Rodriguez, Pau
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (04) : 4071 - 4089
  • [2] Self-supervised Prototype Conditional Few-Shot Object Detection
    Kobayashi, Daisuke
    [J]. IMAGE ANALYSIS AND PROCESSING, ICIAP 2022, PT II, 2022, 13232 : 681 - 692
  • [3] Self-Supervised Approach for Few-shot Hand Gesture Recognition
    Kimura, Naoki
    [J]. ADJUNCT PROCEEDINGS OF THE 35TH ACM SYMPOSIUM ON USER INTERFACE SOFTWARE & TECHNOLOGY, UIST 2022, 2022,
  • [4] Unsupervised Few-Shot Feature Learning via Self-Supervised Training
    Ji, Zilong
    Zou, Xiaolong
    Huang, Tiejun
    Wu, Si
    [J]. FRONTIERS IN COMPUTATIONAL NEUROSCIENCE, 2020, 14
  • [5] A Self-Supervised Deep Learning Framework for Unsupervised Few-Shot Learning and Clustering
    Zhang, Hongjing
    Zhan, Tianyang
    Davidson, Ian
    [J]. PATTERN RECOGNITION LETTERS, 2021, 148 : 75 - 81
  • [6] A Self-Supervised Deep Learning Framework for Unsupervised Few-Shot Learning and Clustering
    Zhang, Hongjing
    Zhan, Tianyang
    Davidson, Ian
    [J]. Pattern Recognition Letters, 2021, 148 : 75 - 81
  • [7] Self-Supervised Tuning for Few-Shot Segmentation
    Zhu, Kai
    Zhai, Wei
    Cao, Yang
    [J]. PROCEEDINGS OF THE TWENTY-NINTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2020, : 1019 - 1025
  • [8] Few-shot object detection based on self-supervised feature pyramid network
    Lv, Wen
    Qi, Xinwei
    Shi, Hongbo
    Tan, Shuai
    Song, Bing
    Tao, Yang
    [J]. JOURNAL OF ELECTRONIC IMAGING, 2024, 33 (02)
  • [9] Self-Supervised Few-Shot Time-Series Segmentation for Activity Recognition
    Xiao, Chunjing
    Chen, Shiming
    Zhou, Fan
    Wu, Jie
    [J]. IEEE TRANSACTIONS ON MOBILE COMPUTING, 2023, 22 (11) : 6770 - 6783
  • [10] FLUID: Few-Shot Self-Supervised Image Deraining
    Rai, Shyam Nandan
    Saluja, Rohit
    Arora, Chetan
    Balasubramanian, Vineeth N.
    Subramanian, Anbumani
    Jawahar, C., V
    [J]. 2022 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2022), 2022, : 418 - 427