A Unified Framework for Image-to-3D Part Generation via Variable Granularity
Paper • 2610.31671 • Published
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("gucci233/FlexPart", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]FlexPart
FlexPart is a unified image-to-3D part generation framework that supports variable-granularity geometric conditions, ranging from sparse points to bounding boxes and segmentation masks, within a single model. It generates separate, assemblable 3D parts from a single image.
This model is presented in the paper A Unified Framework for Image-to-3D Part Generation via Variable Granularity.
The code and usage instructions are available in the official GitHub repository: https://github.com/Gucci233/FlexPart.
- Downloads last month
- 65
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support