Skip to content

Threestudio extension of the paper "Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation".

Notifications You must be signed in to change notification settings

cvlab-kaist/3DFuse-threestudio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

3DFuse - threestudio


This is an threestudio extension of the paper "Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation".

To use it, please install threestudio first and then install this extension in threestudio custom directory.

  • This code is forked from threestudio for VSD backbone version of 3DFuse.
  • For SJC backbone version of 3DFuse, check original 3DFuse repo.

Introduction

We introduce 3DFuse, a novel framework that incorporates 3D awareness into pretrained 2D diffusion models, enhancing the robustness and 3D consistency of score distillation-based methods. For more details, please visit our project page!

Installation

3DFuse - threestudio is extension of threestudio. Before clone this repository, clone and install threestudio first.

Install threestudio environment

This part is the same as original threestudio. Skip it if you already have installed the environment.

❗️CAUTION: YOU DON'T NEED TO CLONE THREESTUDIO REPOSITORY!! THIS PART IS ONLY FOR ENVIRONMENT SETTING.❗️

See installation.md for additional information, including installation via Docker.

  • You must have an NVIDIA graphics card with at least 20GB VRAM and have CUDA installed.
  • Install Python >= 3.8.
  • (Optional, Recommended) Create a virtual environment:
python3 -m virtualenv venv
. venv/bin/activate

# Newer pip versions, e.g. pip-23.x, can be much faster than old versions, e.g. pip-20.x.
# For instance, it caches the wheels of git packages to avoid unnecessarily rebuilding them later.
python3 -m pip install --upgrade pip

Install PyTorch >= 1.12. We have tested on torch1.12.1+cu113 and torch2.0.0+cu118, but other versions should also work fine.

# torch1.12.1+cu113
pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 --extra-index-url https://download.pytorch.org/whl/cu113
# or torch2.0.0+cu118
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
  • (Optional, Recommended) Install ninja to speed up the compilation of CUDA extensions:
pip install ninja
  • Install dependencies:
pip install -r requirements.txt

Install 3DFuse

After installing threestudio, install 3DFuse extension by:

git clone https://ku-cvlab.github.io/3DFuse-threestudio/

cd 3DFuse/RASTER
pip install -e .

Quickstart

python launch.py --config custom/configs/prolificdreamer-fuse.yaml  --train --gpu 0 system.prompt_processor.prompt="a product photo of a toy tank" system.image_dir="your_image_directory"

Acknowledgement

This code is built on the threestudio-project. Thanks to the maintainers for their contribution to the community! Also, we would like to acknowledge the contributions of public projects, including SJC and ControlNet whose code has been utilized in this repository.

Citation

@article{seo2023let,
  title={Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation},
  author={Seo, Junyoung and Jang, Wooseok and Kwak, Min-Seop and Kim, Hyeonsu and Ko, Jaehoon and Kim, Junho and Kim, Jin-Hwa and Lee, Jiyoung and Kim, Seungryong},
  journal={arXiv preprint arXiv:2303.07937},
  year={2023}
}

About

Threestudio extension of the paper "Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation".

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published