Projects per year
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.
Original language | English |
---|---|
Pages (from-to) | 1725–1738 |
Number of pages | 14 |
Journal | International Journal of Machine Learning and Cybernetics |
Volume | 14 |
Issue number | 5 |
Early online date | 2 Dec 2022 |
DOIs | |
Publication status | Published - May 2023 |
Externally published | Yes |
Bibliographical note
Funding Information:This work was supported in part by the National Natural Science Foundation of China under Grants 62202175, 61876066, 62176160, and 61672443, the 67th Chinese Postdoctoral Science Foundation (2020M672631), the Hong Kong RGC General Research Funds under Grant 9042489 (CityU 11206317), Grant 9042816 (CityU 11209819) and Grant 9042322 (CityU 11200116), Natural Science Foundation of Guangdong Province of China (2022A1515010791), Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA), and Natural Science Foundation of Shenzhen (20200804193857002).
Funding Information:
This work was supported in part by the National Natural Science Foundation of China under Grants 62202175, 61876066, 62176160, and 61672443, the 67th Chinese Postdoctoral Science Foundation (2020M672631), the Hong Kong RGC General Research Funds under Grant 9042489 (CityU 11206317), Grant 9042816 (CityU 11209819) and Grant 9042322 (CityU 11200116), Natural Science Foundation of Guangdong Province of China (2022A1515010791), Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA), and Natural Science Foundation of Shenzhen (20200804193857002).
Publisher Copyright:
© 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.
Keywords
- Face retrieval
- Generative adversarial
- Inpainting
- Occlusion
Fingerprint
Dive into the research topics of 'Generative face inpainting hashing for occluded face retrieval'. Together they form a unique fingerprint.Projects
- 1 Active
-
Adaptive Dynamic Range Enhancement Oriented to High Dynamic Display (面向高動態顯示的自適應動態範圍增強)
KWONG, S. T. W., KUO, C. J., WANG, S. & ZHANG, X.
Research Grants Council (HKSAR)
1/01/21 → 30/06/24
Project: Grant Research