---
task_categories:
- image-to-image
tags:
- color-transfer
---
# Color Transfer Dataset
In my publication titled "A Software Test Bed for Sharing and Evaluating Color Transfer Algorithms for Images and 3D Objects", I introduced a dataset comprising a diverse collection of images. These images span various scenes, sizes, and color depths, specifically designed for color transfer evaluation.

## Details
- Representation of five distinct scenes: abstract, city, closeup, interior, and landscape.
- A collection of ten images for each scene, totaling 50 unique images.
- Each of the 50 images is available in four resolutions: 256x256px, 512x512px, 1024x1024px, and 2048x2048px.
- Variation in color depth for all 200 resized images, processed using Photoshop's Indexed Color Mode: 4-color palette, 16-color palette, 256-color palette, and the original color depth.
- A comprehensive compilation of 800 images.
- The initial set of 50 images was sourced from flickr.com, adhering to the CC BY 2.0, CC BY-SA 2.0, and CC BY-NC 2.0 license models. These images were subsequently cropped to a resolution of 2048x2048px.

<p align="center">
  <img src="https://cdn-uploads.huggingface.co/production/uploads/6331b00ab7b8e9d6e89a5dc0/SnGqrxXbphBG4Rg1a9pRw.webp" width="512">
  <em>Figure 1: Overview of all 50 unique images.</em>
</p>

## Citation
If you utilize this dataset in your research, kindly provide a citation:
```
@inproceedings{potechius2023, 
    author = {Potechius, Herbert and Raja, Gunasekaran and Sikora, Thomas and Knorr, Sebastian}, 
    title = {A software test bed for sharing and evaluating color transfer algorithms for images and 3D objects}, 
    year = {2023}, 
    isbn = {9798400704260}, 
    publisher = {Association for Computing Machinery}, 
    address = {New York, NY, USA}, 
    url = {https://doi.org/10.1145/3626495.3626509}, 
    doi = {10.1145/3626495.3626509}, 
    booktitle = {Proceedings of the 20th ACM SIGGRAPH European Conference on Visual Media Production}, 
    articleno = {9}, 
    numpages = {10}, 
    location = {, London, United Kingdom, }, 
    series = {CVMP '23} ,
    keywords={mypublications}
}
```