You're currently viewing an old version of this dataset. To see the current version, click here.

FCSR-GAN: End-to-end learning for joint face completion and super-resolution

The proposed FCSR-GAN uses compound generator and carefully designed losses (adversarial loss, perceptual loss, smooth loss, pixel loss, and face parsing loss) to assure the quality of the recovered face images.

Data and Resources

Cite this as

Jiancheng Cai, Hu Han, Shiguang Shan, Xilin Chen (2025). Dataset: FCSR-GAN: End-to-end learning for joint face completion and super-resolution. https://doi.org/10.57702/4qzud31c

DOI retrieved: January 2, 2025

Additional Info

Field Value
Created January 2, 2025
Last update January 2, 2025
Defined In https://doi.org/10.48550/arXiv.1911.01045
Author Jiancheng Cai
More Authors
Hu Han
Shiguang Shan
Xilin Chen
Homepage https://github.com/swordcheng/FCSR-GAN