Research map / Image generation / Image restoration
DSLR-Quality Photos on Mobile Devices With Deep Convolutional Networks
Andrey Ignatov, Nikolay Kobyshev, Radu Timofte, Kenneth Vanhoey, Luc Van Gool
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 color loss to content + texture loss (ablation) | DPED enhancement network (Ours) · DPED test, iPhone to DSLR | PSNR: 19.05 -> 20.08 | Improved |
| add color loss to content + texture loss (ablation) | DPED enhancement network (Ours) · DPED test, iPhone to DSLR | SSIM: 0.9166 -> 0.9201 | Improved |
| add color loss to content + texture loss (ablation) | DPED enhancement network (Ours) · DPED test, BlackBerry to DSLR | PSNR: 19.64 -> 20.07 | Improved |
| add color loss to content + texture loss (ablation) | DPED enhancement network (Ours) · DPED test, Sony to DSLR | PSNR: 21.59 -> 21.81 | Improved |
| use content loss instead of MSE in the texture-loss setup (with color loss) (ablation) | DPED enhancement network (Ours) · DPED test, iPhone to DSLR | SSIM: 0.9125 -> 0.9201 | Improved |
| use content loss instead of MSE in the texture-loss setup (with color loss) (ablation) | DPED enhancement network (Ours) · DPED test, BlackBerry to DSLR | SSIM: 0.9241 -> 0.9328 | Improved |
| use full color + content + texture loss instead of MSE only (ablation) | DPED enhancement network (Ours) · DPED test, iPhone to DSLR | PSNR: 20.56 -> 20.08 | Hurt |
| use full color + content + texture loss instead of MSE only (ablation) | DPED enhancement network (Ours) · DPED test, BlackBerry to DSLR | SSIM: 0.9292 -> 0.9328 | Improved |
| use full color + content + texture loss instead of MSE only (ablation) | DPED enhancement network (Ours) · DPED test, Sony to DSLR | SSIM: 0.9453 -> 0.9437 | Hurt |
| use full color + content + texture loss instead of MSE only (ablation) | DPED enhancement network (Ours) · DPED test, Sony to DSLR | PSNR: 21.35 -> 21.81 | Improved |
| use the proposed GAN-based network instead of Johnson et al. ResNet | photo enhancement · DPED test, iPhone to DSLR | PSNR: 20.32 -> 20.08 | No clear effect |
| use the proposed method instead of Johnson et al. | photo enhancement · DPED test, Sony to DSLR | SSIM: 0.9434 -> 0.9437 | Improved |
Closest papers by meaning
- Perceptual Quality Assessment of Smartphone Photography (CVPR 2020 · similarity 0.87)
- Transform Your Smartphone into a DSLR Camera: Learning the ISP in the Wild (ECCV 2022 · similarity 0.87)
- A High-Quality Denoising Dataset for Smartphone Cameras (CVPR 2018 · similarity 0.86)
- Real-World Mobile Image Denoising Dataset with Efficient Baselines (CVPR 2024 · similarity 0.86)
- Iterative Residual CNNs for Burst Photography Applications (CVPR 2019 · similarity 0.86)
- DiffRAW: Leveraging Diffusion Model to Generate DSLR-Comparable Perceptual Quality sRGB from Smartphone RAW Images (AAAI 2024 · 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.
