# FCSR-GAN: Joint Face Completion and Super-resolution via Multi-task Learning

arXiv: https://arxiv.org/abs/1911.01045 · HTML: https://mlsanity.com/paper/1911.01045/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Nov 2019.

## Built on by (most cited first, 7 of 7)

- [Task Transformer Network for Joint MRI Reconstruction and Super-Resolution](https://mlsanity.com/paper/2106.06742/) (2021)
- [Deep Learning-based Face Super-Resolution: A Survey](https://mlsanity.com/paper/2101.03749/) (2021)
- [CTCNet: A CNN-Transformer Cooperation Network for Face Image Super-Resolution](https://mlsanity.com/paper/2204.08696/) (2022)
- [Survey on Deep Face Restoration: From Non-blind to Blind and Beyond](https://mlsanity.com/paper/2309.15490/) (2023)
- [Contrastive Attention Network with Dense Field Estimation for Face Completion](https://mlsanity.com/paper/2112.10310/) (2021)
- [Pro-UIGAN: Progressive Face Hallucination from Occluded Thumbnails](https://mlsanity.com/paper/2108.00602/) (2021)
- [When Diffusion Models Forget Who You Are: Identity Preservation in Face Inpainting under Large Occlusions](https://mlsanity.com/paper/2608.04820/) (2026)

## Abstract (arXiv)

Combined variations containing low-resolution and occlusion often present in face images in the wild, e.g., under the scenario of video surveillance. While most of the existing face image recovery approaches can handle only one type of variation per model, in this work, we propose a deep generative adversarial network (FCSR-GAN) for performing joint face completion and face super-resolution via multi-task learning. The generator of FCSR-GAN aims to recover a high-resolution face image without occlusion given an input low-resolution face image with occlusion. The discriminator of FCSR-GAN uses a set of carefully designed losses (an adversarial loss, a perceptual loss, a pixel loss, a smooth loss, a style loss, and a face prior loss) to assure the high quality of the recovered high-resolution face images without occlusion. The whole network of FCSR-GAN can be trained end-to-end using our two-stage training strategy. Experimental results on the public-domain CelebA and Helen databases show that the proposed approach outperforms the state-of-the-art methods in jointly performing face super-resolution (up to 8 ×) and face completion, and shows good generalization ability in cross-database testing. Our FCSR-GAN is also useful for improving face identification performance when there are low-resolution and occlusion in face images.