One Model For All: Partial Diffusion for Unified Try-On and Try-Off in Any Pose
Paper • 2508.04559 • Published
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("liujx233/OneModelForAll", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Check out more details on our project page.
One Model For All : Unified Try-On and Try-Off in Any Pose via LLM-Inspired Bidirectional Tweedie Diffusion
This repository provides the open-source weights for the paper "One Model For All: Unified Try-On and Try-Off in Any Pose via LLM-Inspired Bidirectional Tweedie Diffusion".
Our project supports virtual try-on, virtual try-off, and arbitrary-pose try-on in a unified framework, enabling flexible outfit transfer from only a single portrait and a target garment.
Released Weights
We release two sets of weights for inference on the VITON-HD and DeepFashion-MultiModal datasets, respectively.
Both support high-resolution image generation at [768, 1024].
License
The codes and checkpoints in this repository are under the CC BY-NC-SA 4.0 license.
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support