Research map / Image generation / Image restoration
Needle-Match: Reliable Patch Matching Under High Uncertainty
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 BM3D's regular patch similarity with Needle-based similarity (ablation) | BM3D · Natural images [21], Gaussian noise σ=15, 35, 55, 75 | PSNR (dB): 31.04 -> 31.19; 27.34 -> 27.57; 24.96 -> 25.38; 24.13 -> 24.94 | Helped |
| Replace original-scale patch matching with Needle matching (ablation) | Patch matching for signal-fit evaluation · Noise, σ=35 | RMSE: 22.529 -> 13.89 | Helped |
| Replace original-scale patch matching with Needle matching (ablation) | Patch matching for signal-fit evaluation · Blur | RMSE: 27.484 -> 12.266 | Helped |
| Replace original-scale patch matching with Needle matching (ablation) | Patch matching for signal-fit evaluation · Geometric distortion | RMSE: 33.472 -> 24.219 | Helped |
| Use Needle matching instead of regular patch matching (ablation) | Patch matching without degradation · Clean images, σN=σB=σG=0 | SignalMatchError | Hurt |
| Use simultaneous multiscale matching instead of sequential coarse-to-fine matching (ablation) | Patch matching under severe degradation · Severe blur correspondence estimation | Displacement error | Helped |
| Use unrestricted global regular-patch search instead of local patch search (ablation) | BM3D · Natural images [21], Gaussian noise σ=15, 35, 55, 75 | PSNR (dB): 31.04 -> 26.91; 27.34 -> 26.17; 24.96 -> 24.83; 24.13 -> 23.67 | Hurt |
Closest papers by meaning
- Patched Denoising Diffusion Models For High-Resolution Image Synthesis (ICLR 2024 · similarity 0.86)
- Coarse-to-Fine Embedded PatchMatch and Multi-Scale Dynamic Aggregation for Reference-Based Super-resolution (AAAI 2022 · similarity 0.86)
- From Patches to Images: A Nonparametric Generative Model (ICML 2017 · similarity 0.85)
- Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation (CVPR 2026 · similarity 0.85)
- BUFF: Bayesian Uncertainty Guided Diffusion Probabilistic Model for Single Image Super-Resolution (AAAI 2025 · similarity 0.83)
- Improving Progressive Generation with Decomposable Flow Matching (NeurIPS 2025 · similarity 0.83)
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.