Papers › A U-Net Based Discriminator for Generative Adversarial Networks
A U-Net Based Discriminator for Generative Adversarial Networks
Edgar Schönfeld, Bernt Schiele, Anna Khoreva
Among the major remaining challenges for generative adversarial networks (GANs) is the capacity to synthesize globally and locally coherent images with object shapes and textures indistinguishable from real images. To target this issue we propose an alternative U-Net based discriminator architecture, borrowing the insights from the segmentation literature. The proposed U-Net based architecture allows to provide detailed per-pixel feedback to the generator while maintaining the global coherence of synthesized images, by providing the global image feedback as well. Empowered by the per-pixel response of the discriminator, we further propose a per-pixel consistency regularization technique based on the CutMix data augmentation, encouraging the U-Net discriminator to focus more on semantic and structural changes between real and fake images. This improves the U-Net discriminator training, further enhancing the quality of generated samples. The novel discriminator improves over the state of the art in terms of the standard distribution and image quality metrics, enabling the generator to synthesize images with varying structure, appearance and levels of detail, maintaining global and local realism. Compared to the BigGAN baseline, we achieve an average improvement of 2.7 FID points across FFHQ, CelebA, and the newly introduced COCO-Animals dataset. The code is available at https://github.com/boschresearch/unetgan.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Conditional Image Generation | COCO-Animals | U-Net GAN | FID | 13.73 | #1 of 2 | Archive leaderboard | report |
| Conditional Image Generation | COCO-Animals | U-Net GAN | IS | 12.29 | #1 of 2 | Archive leaderboard | report |
| Conditional Image Generation | COCO-Animals | BigGAN | FID | 16.37 | #2 of 2 | Archive leaderboard | report |
| Conditional Image Generation | COCO-Animals | BigGAN | IS | 11.77 | #2 of 2 | Archive leaderboard | report |
| Image Generation | CelebA 128x128 | U-Net GAN | FID | 2.95 | #1 of 5 | Archive leaderboard | report |
| Image Generation | CelebA 128x128 | U-Net GAN | Inception score | 3.43 | #1 of 5 | Archive leaderboard | report |
| Image Generation | CelebA-HQ 128x128 | U-Net GAN | FID | 2.03 | #1 of 7 | Archive leaderboard | report |
| Image Generation | CelebA-HQ 128x128 | U-Net GAN | Inception score | 3.33 | #1 of 7 | Archive leaderboard | report |
| Image Generation | FFHQ 256 x 256 | U-Net GAN | FID | 7.48 | #30 of 51 | Archive leaderboard | report |
| Image Generation | FFHQ 256 x 256 | BigGAN | FID | 11.48 | #36 of 51 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections