Abstract
In this paper, we propose an effective face completion algorithm using a deep generative model. Different from well-studied background completion, the face completion task is more challenging as it often requires to generate semantically new pixels for the missing key components (e.g., eyes and mouths) that contain large appearance variations. Unlike existing nonparametric algorithms that search for patches to synthesize, our algorithm directly generates contents for missing regions based on a neural network. The model is trained with a combination of a reconstruction loss, two adversarial losses and a semantic parsing loss, which ensures pixel faithfulness and local-global contents consistency. With extensive experimental results, we demonstrate qualitatively and quantitatively that our model is able to deal with a large area of missing pixels in arbitrary shapes and generate realistic face completion results.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 5892-5900 |
| Number of pages | 9 |
| ISBN (Electronic) | 9781538604571 |
| DOIs | |
| Publication status | Published - 2017 Nov 6 |
| Event | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 - Honolulu, United States Duration: 2017 Jul 21 → 2017 Jul 26 |
Publication series
| Name | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
|---|---|
| Volume | 2017-January |
Other
| Other | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
|---|---|
| Country | United States |
| City | Honolulu |
| Period | 17/7/21 → 17/7/26 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Signal Processing
- Computer Vision and Pattern Recognition
Cite this
}
Generative face completion. / Li, Yijun; Liu, Sifei; Yang, Jimei; Yang, Ming Hsuan.
Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 5892-5900 (Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017; Vol. 2017-January).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Generative face completion
AU - Li, Yijun
AU - Liu, Sifei
AU - Yang, Jimei
AU - Yang, Ming Hsuan
PY - 2017/11/6
Y1 - 2017/11/6
N2 - In this paper, we propose an effective face completion algorithm using a deep generative model. Different from well-studied background completion, the face completion task is more challenging as it often requires to generate semantically new pixels for the missing key components (e.g., eyes and mouths) that contain large appearance variations. Unlike existing nonparametric algorithms that search for patches to synthesize, our algorithm directly generates contents for missing regions based on a neural network. The model is trained with a combination of a reconstruction loss, two adversarial losses and a semantic parsing loss, which ensures pixel faithfulness and local-global contents consistency. With extensive experimental results, we demonstrate qualitatively and quantitatively that our model is able to deal with a large area of missing pixels in arbitrary shapes and generate realistic face completion results.
AB - In this paper, we propose an effective face completion algorithm using a deep generative model. Different from well-studied background completion, the face completion task is more challenging as it often requires to generate semantically new pixels for the missing key components (e.g., eyes and mouths) that contain large appearance variations. Unlike existing nonparametric algorithms that search for patches to synthesize, our algorithm directly generates contents for missing regions based on a neural network. The model is trained with a combination of a reconstruction loss, two adversarial losses and a semantic parsing loss, which ensures pixel faithfulness and local-global contents consistency. With extensive experimental results, we demonstrate qualitatively and quantitatively that our model is able to deal with a large area of missing pixels in arbitrary shapes and generate realistic face completion results.
UR - http://www.scopus.com/inward/record.url?scp=85044184136&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044184136&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2017.624
DO - 10.1109/CVPR.2017.624
M3 - Conference contribution
AN - SCOPUS:85044184136
T3 - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
SP - 5892
EP - 5900
BT - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
PB - Institute of Electrical and Electronics Engineers Inc.
ER -