FaceQgen: Semi-Supervised Deep Learning for Face Image Quality Assessment

被引:4
作者
Hernandez-Ortega, Javier [1 ]
Fierrez, Julian [1 ]
Serna, Ignacio [1 ]
Morales, Aythami [1 ]
机构
[1] Univ Autonoma Madrid, Sch Engn, Madrid, Spain
来源
2021 16TH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION (FG 2021) | 2021年
基金
欧盟地平线“2020”;
关键词
D O I
10.1109/FG52635.2021.9667060
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper we develop FaceQgen(1), a No-Reference Quality Assessment approach for face images based on a Generative Adversarial Network that generates a scalar quality measure related with the face recognition accuracy. FaceQgen does not require labelled quality measures for training. It is trained from scratch using the SCface database. FaceQgen applies image restoration to a face image of unknown quality, transforming it into a canonical high quality image, i.e., frontal pose, homogeneous background, etc. The quality estimation is built as the similarity between the original and the restored images, since low quality images experience bigger changes due to restoration. We compare three different numerical quality measures: a) the MSE between the original and the restored images, b) their SSIM, and c) the output score of the Discriminator of the GAN. The results demonstrate that FaceQgen's quality measures are good estimators of face recognition accuracy. Our experiments include a comparison with other quality assessment methods designed for faces and for general images, in order to position FaceQgen in the state of the art. This comparison shows that, even though FaceQgen does not surpass the best existing face quality assessment methods in terms of face recognition accuracy prediction, it achieves good enough results to demonstrate the potential of semi-supervised learning approaches for quality estimation (in particular, data-driven learning based on a single high quality image per subject), having the capacity to improve its performance in the future with adequate refinement of the model and the significant advantage over competing methods of not needing quality labels for its development. This makes FaceQgen flexible and scalable without expensive data curation.
引用
收藏
页数:8
相关论文
共 50 条
  • [21] Improved Active Deep Learning for Semi-Supervised Classification of Hyperspectral Image
    Wang, Qingyan
    Chen, Meng
    Zhang, Junping
    Kang, Shouqiang
    Wang, Yujing
    REMOTE SENSING, 2022, 14 (01)
  • [22] Semi-supervised image depth prediction with deep learning and binocular algorithms
    Tseng, Kuo-Kun
    Zhang, Yaqi
    Zhu, Qinglin
    Yung, K. L.
    Ip, W. H.
    APPLIED SOFT COMPUTING, 2020, 92
  • [23] Semi-supervised Deep Convolutional Transform Learning for Hyperspectral Image Classification
    Singh, Shikha
    Majumdar, Angshul
    Chouzenoux, Emilie
    Chierchia, Giovanni
    2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2022, : 206 - 210
  • [24] Deep Semi-Supervised Learning for Low-Light Image Enhancement
    Qiao, Zhuocheng
    Xu, Wei
    Sun, Li
    Qiu, Song
    Guo, Haoming
    2021 14TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI 2021), 2021,
  • [25] Semi-Supervised Learning With Deep Embedded Clustering for Image Classification and Segmentation
    Enguehard, Joseph
    O'Halloran, Peter
    Gholipour, Ali
    IEEE ACCESS, 2019, 7 : 11093 - 11104
  • [26] An Error Self-learning Semi-supervised Method for No-reference Image Quality Assessment
    Feng, Yingjie
    Li, Sumei
    Hao, Sihan
    2021 INTERNATIONAL CONFERENCE ON VISUAL COMMUNICATIONS AND IMAGE PROCESSING (VCIP), 2021,
  • [27] Theme-Aware Semi-Supervised Image Aesthetic Quality Assessment
    Zhang, Xiaodan
    Zhang, Xun
    Xiao, Yuan
    Liu, Gang
    MATHEMATICS, 2022, 10 (15)
  • [28] Blind Image Quality Assessment using Semi-supervised Rectifier Networks
    Tang, Huixuan
    Joshi, Neel
    Kapoor, Ashish
    2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2014, : 2877 - 2884
  • [29] Image Recognition and Analysis of Intrauterine Residues Based on Deep Learning and Semi-Supervised Learning
    Tao, Tao
    Liu, Kan
    Wang, Li
    Wu, Haiying
    IEEE ACCESS, 2020, 8 : 162785 - 162799
  • [30] Deep semi-supervised learning with contrastive learning and partial label propagation for image data
    Gan, Yanglan
    Zhu, Huichun
    Guo, Wenjing
    Xu, Guangwei
    Zou, Guobing
    KNOWLEDGE-BASED SYSTEMS, 2022, 245