Research map / Image generation / Image restoration
Exploring and Evaluating Image Restoration Potential in Dynamic Scenes
Cheng Zhang, Shaolin Su, Yu Zhu, Qingsen Yan, Jinqiu Sun, Yanning Zhang
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 |
|---|---|---|---|
| Use the proposed degradation-distilling, selectively fused predictor instead of KonCept512 | IRP prediction · DS-IRP | Overall PLCC: 0.7839 -> 0.8687 | Improved |
| Select frames using predicted IRP rather than KonCept512 | Frame filtering followed by DMPHN restoration · GoPro test | PSNR: 31.849 -> 32.140 | Improved |
| Select frames using predicted IRP rather than KonCept512 | Frame filtering · GoPro test | Best-frame selection accuracy: 49/105 -> 57/105 | Improved |
| Add IRP features as auxiliary restoration guidance | CBD denoising model · SIDD | PSNR: 40.227 -> 40.771 | Improved |
| Add IRP features as auxiliary restoration guidance | CBD denoising model · SIDD | SSIM: 0.9793 -> 0.9821 | Improved |
| Add the illumination branch (ablation) | IRP predictor · DS-IRP | Overall PLCC: 0.8543 -> 0.8687 | Improved |
| Add the blur branch (ablation) | IRP predictor · DS-IRP | Overall PLCC: 0.8188 -> 0.8687 | Improved |
| Add the noise branch (ablation) | IRP predictor · DS-IRP | Overall PLCC: 0.8242 -> 0.8687 | Improved |
| Add selective feature fusion (ablation) | IRP predictor · DS-IRP | Overall PLCC: 0.8476 -> 0.8687 | Improved |
| Replace the ResNet50 baseline with the full IRP architecture (ablation) | IRP predictor · DS-IRP | Overall PLCC: 0.7758 -> 0.8687 | Improved |
Closest papers by meaning
- IR-CM: The Fast and General-purpose Image Restoration Method Based on Consistency Model (NeurIPS 2024 · similarity 0.86)
- A Modular Conditional Diffusion Framework for Image Reconstruction (NeurIPS 2024 · similarity 0.86)
- IRPO: Boosting Image Restoration via Post-training GRPO (NeurIPS 2026 · similarity 0.85)
- ReFIR: Grounding Large Restoration Models with Retrieval Augmentation (NeurIPS 2024 · similarity 0.85)
- Reconciling Stochastic and Deterministic Strategies for Zero-shot Image Restoration using Diffusion Model in Dual (CVPR 2025 · similarity 0.85)
- AdaIR: Adaptive All-in-One Image Restoration via Frequency Mining and Modulation (ICLR 2025 · similarity 0.85)
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.
