Learning Pixel Perception for Identity and Illumination Consistency Face Frontalization in the Wild
Summary :
A frontal and realistic face image was synthesized from a single profile face image. It has a wide range of applications in face recognition. Although the frontal face method based on deep learning has made substantial progress in recent years, there is still no guarantee that the generated face has identity consistency and illumination consistency in a significant posture. This paper proposes a novel pixel-based feature regression generative adversarial network (PFR-GAN), which can learn to recover local high-frequency details and preserve identity and illumination frontal face images in an uncontrolled environment. We first propose a Reslu block to obtain richer feature representation and improve the convergence speed of training. We then introduce a feature conversion module to reduce the artifacts caused by face rotation discrepancy, enhance image generation quality, and preserve more high-frequency details of the profile image. We also construct a 30,000 face pose dataset to learn about various uncontrolled field environments. Our dataset includes ages of different races and wild backgrounds, allowing us to handle other datasets and obtain better results. Finally, we introduce a discriminator used for recovering the facial structure of the frontal face images. Quantitative and qualitative experimental results show our PFR-GAN can generate high-quality and high-fidelity frontal face images, and our results are better than the state-of-art results.
- Publication
- IEICE TRANSACTIONS on Information Vol.E106-D No.5 pp.794-803
- Publication Date
- 2023/05/01
- Publicized
- 2022/06/21
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2022DLP0055
- Type of Manuscript
- Special Section PAPER (Special Section on Deep Learning Technologies: Architecture, Optimization, Techniques, and Applications)
- Category
- Person Image Generation
Authors
Yongtang BAO
Shandong University of Science and Technology
Pengfei ZHOU
Shandong University of Science and Technology
Yue QI
Beihang University
Zhihui WANG
Shandong University of Science and Technology
Qing FAN
MiningLamp Technology
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Yongtang BAO, Pengfei ZHOU, Yue QI, Zhihui WANG, Qing FAN, "Learning Pixel Perception for Identity and Illumination Consistency Face Frontalization in the Wild" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 5, pp. 794-803, May 2023, doi: 10.1587/transinf.2022DLP0055.
Abstract: A frontal and realistic face image was synthesized from a single profile face image. It has a wide range of applications in face recognition. Although the frontal face method based on deep learning has made substantial progress in recent years, there is still no guarantee that the generated face has identity consistency and illumination consistency in a significant posture. This paper proposes a novel pixel-based feature regression generative adversarial network (PFR-GAN), which can learn to recover local high-frequency details and preserve identity and illumination frontal face images in an uncontrolled environment. We first propose a Reslu block to obtain richer feature representation and improve the convergence speed of training. We then introduce a feature conversion module to reduce the artifacts caused by face rotation discrepancy, enhance image generation quality, and preserve more high-frequency details of the profile image. We also construct a 30,000 face pose dataset to learn about various uncontrolled field environments. Our dataset includes ages of different races and wild backgrounds, allowing us to handle other datasets and obtain better results. Finally, we introduce a discriminator used for recovering the facial structure of the frontal face images. Quantitative and qualitative experimental results show our PFR-GAN can generate high-quality and high-fidelity frontal face images, and our results are better than the state-of-art results.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022DLP0055/_p
Copy
@ARTICLE{e106-d_5_794,
author={Yongtang BAO, Pengfei ZHOU, Yue QI, Zhihui WANG, Qing FAN, },
journal={IEICE TRANSACTIONS on Information},
title={Learning Pixel Perception for Identity and Illumination Consistency Face Frontalization in the Wild},
year={2023},
volume={E106-D},
number={5},
pages={794-803},
abstract={A frontal and realistic face image was synthesized from a single profile face image. It has a wide range of applications in face recognition. Although the frontal face method based on deep learning has made substantial progress in recent years, there is still no guarantee that the generated face has identity consistency and illumination consistency in a significant posture. This paper proposes a novel pixel-based feature regression generative adversarial network (PFR-GAN), which can learn to recover local high-frequency details and preserve identity and illumination frontal face images in an uncontrolled environment. We first propose a Reslu block to obtain richer feature representation and improve the convergence speed of training. We then introduce a feature conversion module to reduce the artifacts caused by face rotation discrepancy, enhance image generation quality, and preserve more high-frequency details of the profile image. We also construct a 30,000 face pose dataset to learn about various uncontrolled field environments. Our dataset includes ages of different races and wild backgrounds, allowing us to handle other datasets and obtain better results. Finally, we introduce a discriminator used for recovering the facial structure of the frontal face images. Quantitative and qualitative experimental results show our PFR-GAN can generate high-quality and high-fidelity frontal face images, and our results are better than the state-of-art results.},
keywords={},
doi={10.1587/transinf.2022DLP0055},
ISSN={1745-1361},
month={May},}
Copy
TY - JOUR
TI - Learning Pixel Perception for Identity and Illumination Consistency Face Frontalization in the Wild
T2 - IEICE TRANSACTIONS on Information
SP - 794
EP - 803
AU - Yongtang BAO
AU - Pengfei ZHOU
AU - Yue QI
AU - Zhihui WANG
AU - Qing FAN
PY - 2023
DO - 10.1587/transinf.2022DLP0055
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2023
AB - A frontal and realistic face image was synthesized from a single profile face image. It has a wide range of applications in face recognition. Although the frontal face method based on deep learning has made substantial progress in recent years, there is still no guarantee that the generated face has identity consistency and illumination consistency in a significant posture. This paper proposes a novel pixel-based feature regression generative adversarial network (PFR-GAN), which can learn to recover local high-frequency details and preserve identity and illumination frontal face images in an uncontrolled environment. We first propose a Reslu block to obtain richer feature representation and improve the convergence speed of training. We then introduce a feature conversion module to reduce the artifacts caused by face rotation discrepancy, enhance image generation quality, and preserve more high-frequency details of the profile image. We also construct a 30,000 face pose dataset to learn about various uncontrolled field environments. Our dataset includes ages of different races and wild backgrounds, allowing us to handle other datasets and obtain better results. Finally, we introduce a discriminator used for recovering the facial structure of the frontal face images. Quantitative and qualitative experimental results show our PFR-GAN can generate high-quality and high-fidelity frontal face images, and our results are better than the state-of-art results.
ER -
English
