wgsxm/PartCrafter
A latent diffusion transformer model that jointly generates multiple 3D mesh parts and objects from a single RGB image.
Not currently ranked — collecting fresh signals.
star history
PartCrafter is a structured 3D generative model that employs compositional latent diffusion transformers to generate 3D meshes from image inputs. The model jointly synthesizes multiple parts and complete objects in a single forward pass, enabling image-to-3D reconstruction with structural decomposition. The official implementation includes training code, inference scripts, and pre-trained model weights hosted on HuggingFace.
Frequently asked
- What is wgsxm/PartCrafter?
- A latent diffusion transformer model that jointly generates multiple 3D mesh parts and objects from a single RGB image.
- Is PartCrafter open source?
- Yes — wgsxm/PartCrafter is open source, released under the MIT license.
- What language is PartCrafter written in?
- wgsxm/PartCrafter is primarily written in Python.
- How popular is PartCrafter?
- wgsxm/PartCrafter has 2.5k stars on GitHub.
- Where can I find PartCrafter?
- wgsxm/PartCrafter is on GitHub at https://github.com/wgsxm/PartCrafter.