guochengqian/Magic123
PyTorch implementation of a diffusion-based system that generates high-quality 3D objects from single 2D images.
Not currently ranked — collecting fresh signals.
star history
Magic123 is a research project that generates 3D objects from a single image by combining 2D and 3D diffusion priors. The system uses Stable Diffusion for 2D guidance and Zero123 for 3D consistency, enabling higher quality and more 3D-consistent reconstructions compared to methods using only one prior. It is designed for single-image-to-3D conversion, an important task in AI content creation.
Frequently asked
- What is guochengqian/Magic123?
- PyTorch implementation of a diffusion-based system that generates high-quality 3D objects from single 2D images.
- Is Magic123 open source?
- Yes — guochengqian/Magic123 is open source, released under the Apache-2.0 license.
- What language is Magic123 written in?
- guochengqian/Magic123 is primarily written in Jupyter Notebook.
- How popular is Magic123?
- guochengqian/Magic123 has 1.6k stars on GitHub.
- Where can I find Magic123?
- guochengqian/Magic123 is on GitHub at https://github.com/guochengqian/Magic123.