Research map / Image generation / Super-resolution
Event-Guided Super-Resolving Blurry Image via Asymmetric Integral Driven Consistency
Chi Zhang, Xiang Zhang, Lei Yu, Gui-Song Xia, Yuming Fang, Wenhan Yang
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 sharp-to-blurry intensity loss L_S-C to consistency loss (ablation) | SE-SRB · REDS 4x | PSNR: 6.97 -> 21.22 | Improved |
| add reblurring consistency loss L_B-C (ablation) | SE-SRB · REDS 4x | PSNR: 21.22 -> 21.78 | Improved |
| add SR identity loss L_S-I (ablation) | SE-SRB · REDS 4x | PSNR: 21.78 -> 23.81 | Improved |
| add HR consistency loss L_L-C without HR intensity loss (ablation) | SE-SRB · REDS 4x | PSNR: 23.81 -> 7.96 | Hurt |
| add HR sharp-to-blurry intensity loss L_H-C alongside L_L-C (ablation) | SE-SRB · REDS 4x | PSNR: 7.96 -> 24.04 | Improved |
| add HR sharp-to-blurry intensity loss L_H-C alongside L_L-C (ablation) | SE-SRB · REDS 2x | PSNR: 8.01 -> 25.21 | Improved |
| use only inter-frame consistency L_I-C (ablation) | SE-SRB · REDS 4x | PSNR: 6.97 | Hurt |
| self-supervised training with events vs supervised cascades | SE-SRB · REDS 4x | PSNR: 22.23 -> 24.04 (vs EF+CAT) | Improved |
| lightweight dual-branch design | SE-SRB · 256x256 input, RTX 3090 | Runtime (ms): 72.2 -> 24.9 (vs EVDI+DASR) | Improved |
| SE-SRB vs MPR+CAT on GoPro 4x SSIM | SE-SRB · GoPro 4x | SSIM: 0.8253 -> 0.8209 | Hurt |
Closest papers by meaning
- Learning to Super-Resolve Blurry Images With Events (TPAMI 2023 · similarity 0.89)
- Exploiting Blurry Representations for Event-guided Video Super-Resolution (AAAI 2026 · similarity 0.88)
- Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning (CVPR 2024 · similarity 0.87)
- Ultralight Polarity-Split Neuromorphic SNN for Event-Stream Super-Resolution (AAAI 2026 · similarity 0.87)
- Boosting Event Stream Super-Resolution with a Recurrent Neural Network (ECCV 2022 · similarity 0.86)
- Efficient Attention-Sharing Information Distillation Transformer for Lightweight Single Image Super-Resolution (AAAI 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.
