Research map / Image generation / Super-resolution
DreamSR: Towards Ultra-High-Resolution Image Super-Resolution via a Receptive-Field Enhanced Diffusion Transformer
Qingji Dong, Hang Dong, Mingqin Chen, Rui Zhang, Yitong Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add dual-branch MM-ControlNet text branch with global+local prompts (vs global prompt only) (ablation) | DreamSR · RealDeg | MUSIQ: 41.6816 -> 45.6234 | Improved |
| add dual-branch MM-ControlNet (vs patch prompt only) (ablation) | DreamSR · RealDeg | MUSIQ: 41.5523 -> 45.6234 | Improved |
| add dual-branch MM-ControlNet (vs global prompt only) (ablation) | DreamSR · RealDeg | MANIQA: 0.3891 -> 0.4574 | Improved |
| add image-to-image (i2i) degradation pipeline to training (ablation) | DreamSR · RealDeg | MUSIQ: 43.7706 -> 45.6234 | Improved |
| add i2i degradation pipeline (ablation) | DreamSR · RealDeg | CLIPIQA+: 0.6355 -> 0.6647 | Improved |
| add Receptive-Field Enhancement training (512 native-HR patches) (ablation) | DreamSR · RealDeg | MUSIQ: 43.5546 -> 45.6234 | Improved |
| add Receptive-Field Enhancement training (ablation) | DreamSR · RealDeg | CLIPIQA+: 0.5975 -> 0.6647 | Improved |
| add Restoration Acceleration LoRA with 1+16 steps (vs 20 steps without) (ablation) | DreamSR · RealDeg | MUSIQ: 37.9976 -> 43.7706 | Improved |
| add Restoration Acceleration LoRA with 1+16 steps (vs 30 steps without) (ablation) | DreamSR · RealDeg | MUSIQ: 38.3402 -> 43.7706 | Improved |
| add Restoration Acceleration LoRA (ablation) | DreamSR · RealDeg | CLIPIQA+: 0.6042 -> 0.6355 | Improved |
| DreamSR vs BSRGAN on reference metric | DreamSR · RealSR | PSNR: 26.3785 -> 20.9305 | Hurt |
| DreamSR vs best baseline on no-reference metric | DreamSR · RealSR | MUSIQ: 69.8241 -> 70.5560 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 3 on most shared results.
- DiT4SR — DiT4SR: Taming Diffusion Transformer for Real-World Image Super-Resolution (ICCV 2025) · better on 1 of 3 shared results
- FaithDiff — FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution (CVPR 2025) · better on 1 of 2 shared results
- SeeSR — SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (CVPR 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Noise-Started One-Step Image Super-Resolution via LR-Conditioned SplitMeanFlow and GAN Refinement (NeurIPS 2026 · similarity 0.90)
- When Latents Forget Pixels: Restoring Fidelity in Diffusion Transformer Super-Resolution (NeurIPS 2026 · similarity 0.90)
- DREAM: Diffusion Rectification and Estimation-Adaptive Models (CVPR 2024 · similarity 0.90)
- SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (CVPR 2024 · similarity 0.89)
- ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidance (AAAI 2025 · similarity 0.89)
- TSD-SR: One-Step Diffusion with Target Score Distillation for Real-World Image Super-Resolution (CVPR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
