Unsupervised Domain Adaptation using Maximum Mean Covariance Discrepancy and Variational Autoencoder

被引:0
|
作者
Barreto, Fabian [1 ]
Sarvaiya, Jignesh [2 ]
Patnaik, Suprava [3 ]
Yadav, Sushilkumar [4 ]
机构
[1] Xavier Inst Engn, Dept Elect & Telecommun, Mumbai, Maharashtra, India
[2] Sardar Vallabhbhai Natl Inst Technol, Dept Elect, Surat, India
[3] Kalinga Inst Ind Technol, Sch Elect, Bhubaneswar, India
[4] Jio Platforms Ltd, Navi Mumbai, India
关键词
Deep learning; domain adaptation; face recognition; maximum mean covariance discrepancy; transfer learning; variational autoencoders; FACE RECOGNITION;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Face Recognition has progressed tremendously from its initial use of holistic learning models to using hand-crafted, shallow, and deep learning models. DeepFace, a nine-layer Deep Convolutional Neural Network (DCNN), reached near-human performance on unconstrained face recognition for the La-beled Faces in the Wild (LFW) dataset. These models performed very well on the benchmark datasets, but their performance sometimes deteriorated for real-world applications. The problem arose when there was a domain shift due to different distribution spaces of the training and testing models. Few researchers looked at Unsupervised Domain Adaptation (UDA) to find the domain-invariant feature spaces. They tried to minimize the domain discrepancy using a static loss of maximum mean discrepancy (MMD). From MMD, the researchers delved into the higher-order statistics of maximum covariance discrepancy (MCD). MMD and MCD were combined to get maximum mean and covariance discrepancy (MMCD), which captured more information than MMD alone. We use a Variational Autoencoder (VAE) with joint mean and covariance discrepancy to offer a solution for domain adaptation. The proposed MMCD-VAE model uses VAE to measure the discrepancy in the spread of variance around the mean value and uses MMCD to measure the directional discrepancy in the variance. Analysis was done using the TinyFace benchmark dataset and the Bollywood Celebrities dataset. Three objective image quality parameters, namely SSIM, pieAPP, and SIFT feature matching, demonstrate the superiority of MMCD-VAE over the conventional KL-VAE model. MMCD-VAE shows an 18 % improvement in SSIM and a remarkable improvement in the perceptual quality of the image over the conventional KL-VAE model.
引用
收藏
页码:883 / 891
页数:9
相关论文
共 50 条
  • [1] Maximum Mean and Covariance Discrepancy for Unsupervised Domain Adaptation
    Wenju Zhang
    Xiang Zhang
    Long Lan
    Zhigang Luo
    [J]. Neural Processing Letters, 2020, 51 : 347 - 366
  • [2] Maximum Mean and Covariance Discrepancy for Unsupervised Domain Adaptation
    Zhang, Wenju
    Zhang, Xiang
    Lan, Long
    Luo, Zhigang
    [J]. NEURAL PROCESSING LETTERS, 2020, 51 (01) : 347 - 366
  • [3] Unsupervised domain adaptation using maximum mean discrepancy optimization for lithology identification
    Chang, Ji
    Li, Jing
    Kang, Yu
    Lv, Wenjun
    Xu, Ting
    Li, Zerui
    Zheng, Wei Xing
    Han, Hongwei
    Liu, Haining
    [J]. GEOPHYSICS, 2021, 86 (02) : ID19 - ID30
  • [4] Weighted and Class-Specific Maximum Mean Discrepancy for Unsupervised Domain Adaptation
    Yan, Hongliang
    Li, Zhetao
    Wang, Qilong
    Li, Peihua
    Xu, Yong
    Zuo, Wangmeng
    [J]. IEEE TRANSACTIONS ON MULTIMEDIA, 2020, 22 (09) : 2420 - 2433
  • [5] Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
    Saito, Kuniaki
    Watanabe, Kohei
    Ushiku, Yoshitaka
    Harada, Tatsuya
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 3723 - 3732
  • [6] Unsupervised domain adaptation with Joint Adversarial Variational AutoEncoder
    Li, Yuze
    Zhang, Yan
    Yang, Chunling
    [J]. KNOWLEDGE-BASED SYSTEMS, 2022, 250
  • [7] Extreme Learning Machine Based on Maximum Weighted Mean Discrepancy for Unsupervised Domain Adaptation
    Si, Yanna
    Pu, Jiexin
    Zang, Shaofei
    Sun, Lifan
    [J]. IEEE ACCESS, 2021, 9 : 2283 - 2293
  • [8] Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation
    Yan, Hongliang
    Ding, Yukang
    Li, Peihua
    Wang, Qilong
    Xu, Yong
    Zuo, Wangmeng
    [J]. 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 945 - 954
  • [9] Maximum Structural Generation Discrepancy for Unsupervised Domain Adaptation
    Xia, Haifeng
    Jing, Taotao
    Ding, Zhengming
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (03) : 3434 - 3445
  • [10] VARIATIONAL AUTOENCODER BASED UNSUPERVISED DOMAIN ADAPTATION FOR SEMANTIC SEGMENTATION
    Li, Zongyao
    Togo, Ren
    Ogawa, Takahiro
    Haseyama, Miki
    [J]. 2020 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2020, : 2426 - 2430