omaralezaby/Retouch_Transfer_Dataset

Dataset Overview

We manually selected 569 RAW images from the MIT5K dataset to include diverse images with different content, locations, lighting conditions, and cameras. Next, we used Adobe Lightroom software to apply 167 different open-access presets created by photographers to each RAW image. We also carefully selected the presets to include a wide variety of styles and edits. Note that we avoid scene-specific presets, such as portrait presets and geometrical edits (cropping, rotation), to ensure high-quality and focus on general retouching. Lastly, we resize the images to 512 for the smallest side to maintain the original aspect ratio. The dataset consists of approximately 100.000 retouched images with a wide style variety, as shown in the example below.

We split the dataset into 508 images for training and 61 images for evaluation and testing. Addtionally, we split the Presets into 146 for training and evaluation and 22 Presets for testing. Neither the images nor the presets used for the testing benchmark are utilized for training.

Dataset Structure

Retouch_Transfer_Dataset
├── Benchmark
│   ├── Test
│   │   ├── Natural               (61 Images)
│   │   └── Presets
│   │       ├── Preset_146        (61 Images)
│   │       ├── ...
│   │       └── Preset_167        (61 Images)
│   │   
│   ├── Test_References
│   │   ├── Natural               (508 Images)
│   │   └── Presets
│   │       ├── Preset_146        (508 Images)
│   │       ├── ...
│   │       └── Preset_167        (508 Images)
│   │
│   └── references_file.txt
│
│
├── Train
│   ├── Natural                   (508 Images)
│   └── Presets
│       ├── Preset_1              (508 Images)
│       ├── ...
│       └── Preset_145            (508 Images)
│
│
└── Validation
    ├── Natural                   (61 Images)
    └── Presets
        ├── Preset_1              (61 Images)
        ├── ...
        └── Preset_145            (61 Images)

Citation

If you find the dataset helpful, please consider citing the following paper.

@article{elezabi2024inretouch,
  title={INRetouch: Context Aware Implicit Neural Representation for Photography Retouching},
  author={Elezabi, Omar and Conde, Marcos V and Wu, Zongwei and Timofte, Radu},
  journal={arXiv preprint arXiv:2412.03848},
  year={2024}
}      

Contacts

For any inquiries contact
Omar Elezabi: omar.elezabi[at] uni-wuerzburg.de
Marcos V. Conde: marcos.conde [at] uni-wuerzburg.de

License

Copyright (c) 2025 Computer Vision Lab, University of Wurzburg

Licensed under CC BY-NC-SA 4.0 (Attribution-NonCommercial-ShareAlike 4.0 International); you may not use this file except in compliance with the License. You may obtain a copy of the License at

https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode

The code is released for academic research use only. For commercial use, please contact Computer Vision Lab, University of Würzburg. Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.