Spaces:
Running
Running
Kandinsky Inpainting Project
This project uses the Kandinsky inpainting pipeline and a mask generation script to perform inpainting on an image. The mask is generated using the YOLOv8s seg model from the Ultralytics library for mask generation. The mask is then inverted and used in the Kandinsky inpainting pipeline
Installation
To install the necessary requirements, you can use pip:
pip install -r requirements.txt
wandb login
huggingface-cli login
cd scripts
This will install all necessary libraries for this project, including PIL, numpy, Ultralytics wandb, diffuser etc.
cd in to scripts
specify/create the folders manually before this
python run.py --image_path /path/to/image.jpg --prompt 'prompt' --negative_prompt 'negative prompt' --output_dir /path/to/output --mask_dir /path/to/mask --uid unique_id
Some Experiments
Here are some of my experiments with the following models