Research map / Architectures and efficiency / CNN design
Gaussian Kernel Mixture Network for Single Image Defocus Deblurring
Yuhui Quan, Zicong Wu, Hui Ji
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 multi-scale (T=3) estimation instead of single scale (ablation) | GKMNet · DPD test set | PSNR (dB): 23.63 -> 25.46 | Improved |
| add Conv-LSTM in the attention prediction unit (ablation) | GKMNet · DPD test set | PSNR (dB): 25.00 -> 25.46 | Improved |
| wrap SRAM in GKM unrolled fixed-point structure instead of using it as a plain end-to-end network (ablation) | GKMNet · DPD test set | PSNR (dB): 21.98 -> 25.46 | Improved |
| learn GCM kernels initialized from Gaussian kernels (ablation) | GKMNet · DPD test set | PSNR (dB): 25.46 -> 25.49 | No clear effect |
| learn GCM kernels with random initialization (ablation) | GKMNet · DPD test set | PSNR (dB): 25.46 -> 25.35 | Hurt |
| use 3 iterations at a single scale (S=1,R=3) instead of S=3,R=1 (ablation) | GKMNet · DPD test set | PSNR (dB): 25.47 -> 25.11 | Hurt |
| add a second iteration per scale (S=3,R=2) (ablation) | GKMNet · DPD test set | PSNR (dB): 25.47 -> 25.53 PSNR; SSIM 0.789 -> 0.773 | Mixed |
| use GKMNet single-image model versus dual-pixel two-view DPDNet-D | GKMNet · DPD test set | PSNR (dB): 25.12 -> 25.47 | Improved |
| use GKMNet instead of AttNet under heavy noise | RTF noise 2.55 · RTF | PSNR (dB): 23.83 -> 25.01 | Improved |
Closest papers by meaning
- Deep Single Image Defocus Deblurring via Gaussian Kernel Mixture Learning (TPAMI 2024 · similarity 0.96)
- Prior and Prediction Inverse Kernel Transformer for Single Image Defocus Deblurring (AAAI 2024 · similarity 0.89)
- Single Image Defocus Deblurring Using Kernel-Sharing Parallel Atrous Convolutions (ICCV 2021 · similarity 0.88)
- Learnable Blur Kernel for Single-Image Defocus Deblurring in the Wild (AAAI 2023 · similarity 0.88)
- DeFusionNET: Defocus Blur Detection via Recurrently Fusing and Refining Discriminative Multi-Scale Deep Features (TPAMI 2020 · similarity 0.87)
- Residual Diffusion Deblurring Model for Single Image Defocus Deblurring (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.
