Occluded Face Recovery by Image Retrieval

In this paper, we propose an occluded face recovery framework to improve the face recognition rate for the occlusion case. The occluded facial image of an unseen testing subject is automatically recovered, the generated results are diverse and fit the personal characteristics of the observable regio...

Full description

Saved in:
Bibliographic Details
Published in2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) pp. 1 - 2
Main Authors Tu, Ching-Ting, Lee, Kuan-Hsin
Format Conference Proceeding
LanguageEnglish
Published IEEE 16.11.2021
Subjects
Communication systems
Data models
Face recognition
Generative adversarial network
Image inpainting
Image recognition
Image retrieval
Signal processing
Testing
Variational Autoencoders
Online AccessGet full text

Cover

More Information
Summary:In this paper, we propose an occluded face recovery framework to improve the face recognition rate for the occlusion case. The occluded facial image of an unseen testing subject is automatically recovered, the generated results are diverse and fit the personal characteristics of the observable regions, and thus can be used to increase the face recognition performance for occlusion cases.
DOI:10.1109/ISPACS51563.2021.9650973