Your browser doesn't support javascript.
Generative face inpainting hashing for occluded face retrieval.
Yang, Yuxiang; Tian, Xing; Ng, Wing W Y; Wang, Ran; Gao, Ying; Kwong, Sam.
  • Yang Y; School of Computer Science and Engineering, South China University of Technology, Guangzhou, 510006 China.
  • Tian X; School of Computer Science and Engineering, South China University of Technology, Guangzhou, 510006 China.
  • Ng WWY; School of Computer Science and Engineering, South China University of Technology, Guangzhou, 510006 China.
  • Wang R; College of Mathematics and Statistics, Shenzhen University, Shenzhen, 518060 China.
  • Gao Y; School of Computer Science and Engineering, South China University of Technology, Guangzhou, 510006 China.
  • Kwong S; Department of Computer Science, City University of Hong Kong, Hongkong, 999077 China.
Int J Mach Learn Cybern ; : 1-14, 2022 Dec 02.
Article in English | MEDLINE | ID: covidwho-2306069
ABSTRACT
COVID-19 has resulted in a significant impact on individual lives, bringing a unique challenge for face retrieval under occlusion. In this paper, an occluded face retrieval method which consists of generator, discriminator, and deep hashing retrieval network is proposed for face retrieval in a large-scale face image dataset under variety of occlusion situations. In the proposed method, occluded face images are firstly reconstructed using a face inpainting model, in which the adversarial loss, reconstruction loss and hash bits loss are combined for training. With the trained model, hash codes of real face images and corresponding reconstructed face images are aimed to be as similar as possible. Then, a deep hashing retrieval network is used to generate compact similarity-preserving hashing codes using reconstructed face images for a better retrieval performance. Experimental results show that the proposed method can successfully generate the reconstructed face images under occlusion. Meanwhile, the proposed deep hashing retrieval network achieves better retrieval performance for occluded face retrieval than existing state-of-the-art deep hashing retrieval methods.
Keywords

Full text: Available Collection: International databases Database: MEDLINE Language: English Journal: Int J Mach Learn Cybern Year: 2022 Document Type: Article

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: International databases Database: MEDLINE Language: English Journal: Int J Mach Learn Cybern Year: 2022 Document Type: Article