Research map / Image generation / Image restoration
Learnable Blur Kernel for Single-Image Defocus Deblurring in the Wild
Jucai Zhai, Pengcheng Zeng, Chihao Ma, Jie Chen, Yong Zhao
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 |
|---|---|---|---|
| guide deblurring with learned defocus maps and use DefocusGAN | single-image defocus deblurring · DPDD Indoor & Outdoor | PSNR: 25.35 -> 25.56 | Improved |
| guide deblurring with learned defocus maps and use DefocusGAN | single-image defocus deblurring · DPDD Indoor & Outdoor | LPIPS: 0.225 -> 0.111 | Improved |
| use the proposed system rather than DP-view-input methods | DPDNet and RDPDNet · DPDD | PSNR: 25.39 -> 25.56 | Improved |
| use a learnable blur kernel instead of a Butterworth kernel (ablation) | defocus-map estimation · DPDD | PSNR: 23.31 -> 23.56 | Improved |
| use a learnable blur kernel instead of a Gaussian kernel (ablation) | defocus-map estimation · DPDD | PSNR: 22.56 -> 23.56 | Improved |
| add RCAB (ablation) | single-scale base network · DPDD | PSNR: 24.73 -> 25.10 | Improved |
| add defocus-map guidance (DGB) (ablation) | Base + RCAB · DPDD | PSNR: 25.10 -> 25.30 | Improved |
| add the multi-scale module (ablation) | Base + RCAB + DGB · DPDD | PSNR: 25.30 -> 25.47 | Improved |
| add perceptual loss (ablation) | Base + RCAB + DGB + MS · DPDD | LPIPS: 0.172 -> 0.115 | Improved |
| add GAN training with defocus adversarial loss (ablation) | DefocusGAN with perceptual loss · DPDD | LPIPS: 0.115 -> 0.111 | Improved |
| replace defocus adversarial loss with doubleGAN loss (ablation) | DefocusGAN · DPDD | PSNR: 25.42 -> 24.92 | Hurt |
| replace defocus adversarial loss with doubleGAN loss (ablation) | DefocusGAN · DPDD | LPIPS: 0.160 -> 0.132 | Improved |
Closest papers by meaning
- Reblurring-Guided Single Image Defocus Deblurring: A Learning Framework with Misaligned Training Pairs (IJCV 2025 · similarity 0.93)
- United Defocus Blur Detection and Deblurring via Adversarial Promoting Learning (ECCV 2022 · similarity 0.93)
- Residual Diffusion Deblurring Model for Single Image Defocus Deblurring (AAAI 2025 · similarity 0.92)
- Prior and Prediction Inverse Kernel Transformer for Single Image Defocus Deblurring (AAAI 2024 · similarity 0.91)
- Quad-Pixel Image Defocus Deblurring: A New Benchmark and Model (CVPR 2025 · similarity 0.91)
- LDP: Language-driven Dual-Pixel Image Defocus Deblurring Network (CVPR 2024 · similarity 0.89)
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.
