We provide a GUI which can run on CPU-only devices as well as devices with a GPU.
Try out iColoriT with your own images and color hints!
Checkpoints for iColoriT models are available in the links below.
| Backbone | Link | |
|---|---|---|
| iColoriT | ViT-B | iColoriT (Google Drive) |
| iColoriT-S | ViT-S | iColoriT-S (Google Drive) |
| iColoriT-T | ViT-Ti | iColoriT-T (Google Drive) |
Our code is implemented in Python 3.8, torch>=1.8.2, and PyQt5
git clone https://github.com/pmh9960/iColoriT.git
pip install -r requirements.txt
Once you have satisfied all the requirements, you can run the base iColoriT model by executing
python3 icolorit_ui.py --model_path <path/to/checkpoint> --target_image <path/to/image>
- Left click on Drawing Pad to select the hint location.
- Left click on the ab Color Gamut to select a color.
- Right click on a hint location to undo the click.
- Press Restart to undo all hints.
- Press Save to save the colorized image.
- Press Load to load another image.
Our GUI is an updated version of the interactive-deep-colorization. Thanks for sharing the codes!