MVDream-diffusers Model Card

This is a port of https://huggingface.co./MVDream/MVDream into diffusers.

And get ported weights from https://huggingface.co./ashawkey/mvdream-sd2.1-diffusers
In ashawkey's work, UNet did not ported to diffusers.

This work has been fully ported to diffusers, including UNet.

Diffusers

import torch
from diffusers import DiffusionPipeline
from diffusers.utils import make_image_grid
from PIL import Image

pipe = DiffusionPipeline.from_pretrained(
    "kiigii/mvdream-diffusers",
    torch_dtype=torch.float16,
    trust_remote_code=True,
)
pipe.to("cude")

prompt = "a cute owl 3d model"

mv_images = pipe(
    prompt=prompt,
    guidance_scale=5,
    num_inference_steps=30,
    elevation=0,
    num_images_per_prompt=1
).images
mv_grid = make_image_grid(mv_images[:4], 2, 2)
mv_grid.save("mv_image.png")

Citation

@article{shi2023MVDream,
  author = {Shi, Yichun and Wang, Peng and Ye, Jianglong and Mai, Long and Li, Kejie and Yang, Xiao},
  title = {MVDream: Multi-view Diffusion for 3D Generation},
  journal = {arXiv:2308.16512},
  year = {2023},
}

Misuse, Malicious Use, and Out-of-Scope Use

The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating images that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes.