Research map / Image generation / Image restoration
Blur-Invariant Deep Learning for Blind-Deblurring
T. M. Nimisha, Akash Kumar Singh, A. N. Rajagopalan
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 residual (ResNet) blocks to the denoising autoencoder (ablation) | Encoder-decoder module · Autoencoder reconstruction of noisy image (Fig. 4) | PSNR (dB): 23.1 dB -> 29.5 dB | Improved |
| train GAN on autoencoder feature space instead of regressing clean pixels directly (ablation) | Blur-invariant GAN generator · Qualitative blind deblurring (Fig. 5) | qualitative output quality | Improved |
| add gradient loss to the MSE loss for autoencoder training (ablation) | Autoencoder · Autoencoder training | edge preservation | Improved |
| use l1 instead of l2 loss for feature back-propagation | GAN feature loss · GAN training | training behavior | Improved |
| replace iterative prior-based deblurring with feed-forward network on GPU | Ours vs Xu [37] · Run time, average image 1024x700 | run time: 34 sec -> 3.4 sec | Improved |
| compare against dark channel prior method on PSNR | Ours vs Pan [23] · Dataset of [29] (80 images, 8 kernels) | PSNR (dB): 31.16 -> 30.54 | Hurt |
Closest papers by meaning
- DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks (CVPR 2018 · similarity 0.92)
- A Neural Approach to Blind Motion Deblurring (ECCV 2016 · similarity 0.92)
- Neural Blind Deconvolution Using Deep Priors (CVPR 2020 · similarity 0.92)
- Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring (CVPR 2024 · similarity 0.90)
- Parameterized Blur Kernel Prior Learning for Local Motion Deblurring (CVPR 2025 · similarity 0.88)
- Real-World Efficient Blind Motion Deblurring via Blur Pixel Discretization (CVPR 2024 · similarity 0.87)
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.
