Research map / Image generation
Super-resolution: research map
771 papers on Super-resolution in Image generation, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 6 approaches. The busiest year so far is 2026.
What it is
Super-resolution turns a low-resolution image into a sharper, higher-resolution one. Real photos are degraded in unknown ways (blur, noise, compression), so much of the work is about realistic degradations, plus efficient architectures and diffusion models that hallucinate plausible detail.
Terms you will see on this page:
- LR / HR
- low-resolution input and high-resolution target.
- Blind super-resolution
- super-resolution when the degradation is unknown.
- Perceptual quality
- how real an output looks to people, as opposed to pixel accuracy (PSNR).
Start reading with the most cited papers here:
- Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network (CVPR 2017 · 12,597 citations)
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (CVPR 2016 · 7,310 citations)
Within Image generation, its share shrank from 5.6% in 2023–24 to 4.0% in 2025–26 (117 → 165 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Super-resolution in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Diffusion-based super-resolution rising: 19% → 38% of the topic (22 → 63 papers)
- Perceptual and real-world SR rising: 20% → 24% of the topic (23 → 39 papers)
- Efficient lightweight super-resolution rising: 23% → 26% of the topic (27 → 43 papers)
- Real-world super-resolution falling: 19% → 7% of the topic (22 → 11 papers)
- CNN and transformer SR architectures falling: 15% → 4% of the topic (17 → 6 papers)
Approaches and key papers
Super-resolution architectures · 610 papers
Keywords: image super, attention, resolution diffusion
Approaches in this cluster:
- Efficient lightweight super-resolution (187 papers)
Reduce super-resolution cost with lightweight networks, selective processing, and mixtures of experts. - Perceptual and real-world SR (154 papers)
Improve realism and semantic accuracy of super-resolution with perceptual losses and semantic guidance. - CNN and transformer SR architectures (153 papers)
Design deep recursive, attention, and transformer architectures for single-image super-resolution. - Diffusion-based super-resolution (116 papers)
Accelerate diffusion super-resolution with one-step distillation and residual shifting.
Most cited and most cited since 2024:
- Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network (CVPR 2017 · 12,597 citations)
- Residual Dense Network for Image Super-Resolution (CVPR 2018 · 4,176 citations)
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step (CVPR 2024 · 167 citations)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (AAAI 2024 · 126 citations)
Real-world and blind super-resolution · 161 papers
Keywords: lr, hr, degradation
Approaches in this cluster:
- Real-world super-resolution (117 papers)
Handle ill-posed super-resolution with degradation modeling, adaptation, and frequency-aware representations. - Blind super-resolution (44 papers)
Estimate or represent unknown degradation kernels for blind super-resolution.
Most cited and most cited since 2024:
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (CVPR 2016 · 7,310 citations)
- Learning a Single Convolutional Super-Resolution Network for Multiple Degradations (CVPR 2018 · 1,151 citations)
- SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (CVPR 2024 · 189 citations)
- SGNet: Structure Guided Network via Gradient-Frequency Awareness for Depth Map Super-resolution (AAAI 2024 · 49 citations)
Related topics in Image generation
- Autoregressive and latent generation (2,100)
- GANs (780)
- Editing and style transfer (1,430)
- Fast diffusion sampling (2,155)
- Image restoration (2,009)
- Text-to-image diffusion (1,916)
