Test2train Amine Face

🚀 Model Card for My First pre-trained model -- test2train_anime_face

This model is a diffusion model designed for unconditional image generation of 64*64 anime-style face pictures. It offers a unique way to generate anime faces, leveraging the power of diffusion models in the field of image synthesis.

🚀 Quick Start

The model uses the anime-faces dataset for training. This dataset contains 21551 anime faces scraped from www.getchu.com and then cropped using the anime face detection algorithm from https://github.com/nagadomi/lbpcascade_animeface.

It's important to note that generating multiple pictures at once may lead to distorted faces. Testing has shown that generating one picture at a time yields the best results and reduces the likelihood of distorted faces.

Here are some sample images generated by the model:

💻 Usage Examples

Basic Usage

from diffusers import DDPMPipeline

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
pipeline = DDPMPipeline.from_pretrained('Chilli-b/test2train_amine_face').to(device)
image = pipeline().images[0]
image

📄 License

This project is licensed under the MIT License.