Research map / Image generation / Autoregressive and latent generation
Towards Universal Computational Aberration Correction in Photographic Cameras: A Comprehensive Benchmark Analysis
Xiaolong Qian, Qi Jiang, Yao Gao, Lei Sun, Zhonghua Yi, Kailun Yang, Luc Van Gool, Kaiwei Wang
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 GAN training (ablation) | FeMaSR · UniCAC | ClipIQA: 0.491 -> 0.512 | Improved |
| add GAN training (ablation) | FeMaSR · UniCAC | LPIPS: 0.147 -> 0.145 | Improved |
| add codebook clean-image prior (ablation) | FeMaSR · UniCAC | LPIPS: 0.145 -> 0.136 | Improved |
| add codebook clean-image prior (ablation) | FeMaSR · UniCAC | PSNR: 26.86 -> 26.94 | Improved |
| add FoV input (ablation) | DFUnet · UniCAC | PSNR: 25.88 -> 26.33 | Improved |
| add FoV input (ablation) | DFUnet · UniCAC | LPIPS: 0.253 -> 0.252 | No clear effect |
| add PSF attention (ablation) | PART · UniCAC | PSNR: 27.22 -> 28.10 | Improved |
| add PSF attention (ablation) | PART · UniCAC | LPIPS: 0.264 -> 0.228 | Improved |
| train on UniCAC (with aspherical lenses) instead of AODLib | FeMaSR · UniCAC test | PSNR: 24.15 -> 26.94 | Improved |
| train on UniCAC instead of AODLib | SwinIR · UniCAC test | PSNR: 25.05 -> 27.22 | Improved |
| train on UniCAC instead of AODLib | FOV-KPN · UniCAC test | PSNR: 23.90 -> 26.34 | Improved |
| increase aberration severity from L1 to L5 | PART · UniCAC | PSNR: 29.22 -> 26.19 | Hurt |
Closest papers by meaning
- Revisiting Cross-Channel Information Transfer for Chromatic Aberration Correction (ICCV 2017 · similarity 0.89)
- Fast Two-Step Blind Optical Aberration Correction (ECCV 2022 · similarity 0.88)
- Correcting Optical Aberration via Depth-Aware Point Spread Functions (TPAMI 2024 · similarity 0.88)
- Universal and Flexible Optical Aberration Correction Using Deep-Prior Based Deconvolution (ICCV 2021 · similarity 0.88)
- A Physics-informed Low-rank Deep Neural Network for Blind and Universal Lens Aberration Correction (CVPR 2024 · similarity 0.86)
- Continuous Optical Zooming: A Benchmark for Arbitrary-Scale Image Super-Resolution in Real World (CVPR 2024 · 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.
