Research map / Image generation / Image restoration
Fast Two-Step Blind Optical Aberration Correction
Thomas Eboli, Jean-Michel Morel, Gabriele Facciolo
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 |
|---|---|---|---|
| use proposed small network instead of prior CNN | aberration correction · 24-megapixel photo (6000x4000), NVIDIA 3090 | runtime (seconds): 30 -> 1.7 | Improved |
| use proposed network with 160K parameters instead of 17 million | aberration correction · 24-megapixel photo (6000x4000), NVIDIA 3090 | compute: 27.3 teraflops -> 33.1 gigaflops | Improved |
| train with red-green and blue-green residual loss instead of plain L1 regression loss (ablation) | lateral chromatic aberration CNN · real raw image | visual color artifacts | Improved |
| use blind Gaussian PSF removal instead of non-parametric blind algorithm of Anger et al. | blind deblurring · 870 synthetic images (400x400) | SSIM ratio R | Improved |
| use CNN edge correction instead of phase correlation, PLK or radial model | lateral chromatic aberration compensation · 1,500 real test patches | energy E | Improved |
Closest papers by meaning
- Correcting Optical Aberration via Depth-Aware Point Spread Functions (TPAMI 2024 · similarity 0.90)
- Towards Universal Computational Aberration Correction in Photographic Cameras: A Comprehensive Benchmark Analysis (CVPR 2026 · similarity 0.88)
- A Physics-informed Low-rank Deep Neural Network for Blind and Universal Lens Aberration Correction (CVPR 2024 · similarity 0.88)
- A Physics-Informed Blur Learning Framework for Imaging Systems (CVPR 2025 · similarity 0.85)
- Real-World Efficient Blind Motion Deblurring via Blur Pixel Discretization (CVPR 2024 · similarity 0.83)
- Dynamic Gaussian Splatting from Defocused and Motion-blurred Monocular Videos (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.
