mlsanityData to 7 Oct 2026

SupResDiffGAN a new approach for the Super-Resolution task

arXiv 2504.13622 ↗First posted 18 Apr 2025eess.IV cs.CVcited by 0

On the map: Diffusion models → Inverse problems and image restoration

Abstract (arXiv)

In this work, we present SupResDiffGAN, a novel hybrid architecture that combines the strengths of Generative Adversarial Networks (GANs) and diffusion models for super-resolution tasks. By leveraging latent space representations and reducing the number of diffusion steps, SupResDiffGAN achieves significantly faster inference times than other diffusion-based super-resolution models while maintaining competitive perceptual quality. To prevent discriminator overfitting, we propose adaptive noise corruption, ensuring a stable balance between the generator and the discriminator during training. Extensive experiments on benchmark datasets show that our approach outperforms traditional diffusion models such as SR3 and I2SB in efficiency and image quality. This work bridges the performance gap between diffusion- and GAN-based methods, laying the foundation for real-time applications of diffusion models in high-resolution image generation.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models. Builds on 32 of its 45 references that are in the map. First posted on arXiv 18 Apr 2025.

Builds on

32 of 45 references.

Closest ancestors · 8

Same direction, newest first

  1. 2024OmniSSR: Zero-shot Omnidirectional Image Super-Resolution using Stable Diffusion Model16 Apr
  2. 2023SinSR: Diffusion-Based Image Super-Resolution in a Single Step23 Nov
  3. 2023Pixel-Aware Stable Diffusion for Realistic Image Super-resolution and Personalized Stylization28 Aug
  4. 2023ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting23 Jul
  5. 2023Exploiting Diffusion Prior for Real-World Image Super-Resolution11 May
  6. 2023Implicit Diffusion Models for Continuous Super-Resolution29 Mar
  7. 2023Denoising Diffusion Probabilistic Models for Robust Image Super-Resolution in the Wild15 Feb
  8. 2021SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Models30 Apr

Other references · 24

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2020Denoising Diffusion Implicit Modelscited by 9,655
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  7. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  8. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  9. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  10. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  11. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,250
  12. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,218
12 more
  1. 2021Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Datacited by 961
  2. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  3. 2021Designing a Practical Degradation Model for Deep Blind Image Super-Resolutioncited by 524
  4. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 515
  5. 2019Toward Real-World Single Image Super-Resolution: A New Benchmark and A New Modelcited by 400
  6. 2022Diffusion-GAN: Training GANs with Diffusioncited by 203
  7. 2023UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANscited by 165
  8. 2020ESRGAN+ : Further Improving Enhanced Super-Resolution Generative Adversarial Networkcited by 27
  9. 2024Structure-Guided Adversarial Training of Diffusion Modelscited by 11
  10. 2024Latent Denoising Diffusion GAN: Faster sampling, Higher image qualitycited by 4
  11. 2024Does Diffusion Beat GAN in Image Super Resolution?cited by 3
  12. 2023DifAugGAN: A Practical Diffusion-style Data Augmentation for GAN-based Single Image Super-resolutioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Improving the Stability and Efficiency of Diffusion Models for Content Consistent Super-Resolution
  2. 2024A Wavelet Diffusion GAN for Image Super-Resolution
  3. 2024Adversarial Diffusion Compression for Real-World Image Super-Resolution
  4. 2026TUDSR: Twice Upsampling-Diffusion for Higher Super-Resolution
  5. 2024AddSR: Accelerating Diffusion-based Blind Super-Resolution with Adversarial Diffusion Distillation
  6. 2026TEASR: Training-Efficient Any-Step Diffusion Transformer for Real-World Image Super-Resolution
  7. 2026Improved Adversarial Diffusion Compression for Real-World Video Super-Resolution
  8. 2024One Step Diffusion-based Super-Resolution with Time-Aware Distillation