Research map / Image generation / Image restoration
Modelling the Scene Dependent Imaging in Cameras With a Deep Neural Network
Seonghyeon Nam, Seon Joo Kim
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 learnable multi-scale color histogram features instead of an MLP color mapping | scene-dependent imaging network · RAW-to-sRGB, Canon 5D Mark III | PSNR mean (dB): 25.43 -> 29.63 | Improved |
| use histogram features instead of local-receptive-field SRCNN | scene-dependent imaging network · RAW-to-sRGB, Canon 5D Mark III | PSNR mean (dB): 27.62 -> 29.63 | Improved |
| use histogram features instead of hypercolumn CNN features | scene-dependent imaging network · RAW-to-sRGB, Nikon D600 | PSNR mean (dB): 27.99 -> 28.85 | Improved |
| use histogram features instead of FCN features | scene-dependent imaging network · RAW-to-sRGB, Samsung Galaxy S7 | PSNR mean (dB): 20.80 -> 30.14 | Improved |
| use histogram features instead of MLP | scene-dependent imaging network · sRGB-to-RAW, Canon 5D Mark III | PSNR mean (dB): 34.72 -> 35.16 | Improved |
| use histogram features instead of MLP | scene-dependent imaging network · sRGB-to-RAW, Samsung Galaxy S7 | PSNR mean (dB): 29.56 -> 31.67 | Improved |
| use scene-aware SRCNN instead of scene-independent MLP | SRCNN vs MLP · sRGB-to-RAW, Canon 5D Mark III | PSNR mean (dB): 34.72 -> 32.34 | Hurt |
Closest papers by meaning
- Model-Based Image Signal Processors via Learnable Dictionaries (AAAI 2022 · similarity 0.84)
- Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks (CVPR 2018 · similarity 0.84)
- Learning sRGB-to-Raw-RGB De-Rendering With Content-Aware Metadata (CVPR 2022 · similarity 0.83)
- Depth CNNs for RGB-D Scene Recognition: Learning from Scratch Better than Transferring from RGB-CNNs (AAAI 2017 · similarity 0.83)
- Self-Calibrating Neural Radiance Fields (ICCV 2021 · similarity 0.83)
- Exploiting Deblurring Networks for Radiance Fields (CVPR 2025 · similarity 0.82)
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.
