Research map / Image generation / Super-resolution
Tiled Prompts: Overcoming Prompt Misguidance in Image and Video Super-Resolution
Bryan Sangwoo Kim, Jonghyun Park, Jong Chul Ye
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 |
|---|---|---|---|
| replace global prompt with tile-specific local prompts | DiT4SR · LSDIR1K | ImageReward: -1.5901 -> -0.6771 | Improved |
| replace global prompt with tile-specific local prompts | DiT4SR · LSDIR1K | NIQE: 2.9427 -> 2.9040 | Improved |
| replace global prompt with tile-specific local prompts | DiT4SR · Urban100 | MUSIQ: 53.1372 -> 54.9203 | Improved |
| replace global prompt with tile-specific local prompts | STAR · VideoLQ | DOVER: 53.5704 -> 54.4158 | Improved |
| replace global prompt with tile-specific local prompts | STAR · MVSR4x | FasterVQA: 0.7402 -> 0.7925 | Improved |
| replace global prompt with tile-specific local prompts | STAR · RealVSR | FAST-VQA: 0.7268 -> 0.7244 | Hurt |
| use only local prompts instead of global+local concatenation (ablation) | STAR · VideoLQ | NIQE: 4.7530 -> 4.5913 | Improved |
| add global prompt to local prompts (Global+Local) (ablation) | DiT4SR · LSDIR1K | CLIP Score: 25.3348 -> 25.7925 | Improved |
| add a text prompt (global) versus none (ablation) | DiT4SR · LSDIR1K | NIQE: 3.4537 -> 2.9427 | Improved |
| add a text prompt (global) versus none (ablation) | DiT4SR · Urban100 | MUSIQ: 56.1273 -> 53.1372 | Hurt |
| use tiled prompts with higher CFG scale (ablation) | STAR · VideoLQ/RealVSR/MVSR | NIQE, CLIPIQA, MUSIQ | Improved |
| apply tiled prompts to OSEDiff backbone | OSEDiff · DIV2K | NIQE: 3.7234 -> 3.6821 | Improved |
Closest papers by meaning
- The Power of Context: How Multimodality Improves Image Super-Resolution (CVPR 2025 · similarity 0.88)
- Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment (NeurIPS 2025 · similarity 0.87)
- SRSR: Enhancing Semantic Accuracy in Real-World Image Super-Resolution with Spatially Re-Focused Text-Conditioning (NeurIPS 2025 · similarity 0.86)
- Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution (CVPR 2024 · similarity 0.86)
- SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (CVPR 2024 · similarity 0.86)
- Prompting Hard or Hardly Prompting: Prompt Inversion for Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.86)
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.
