This repository contains the official implementation of GSGEN: Text-to-3D using Gaussian Splattng.
https://github.com/gsgen3d/gsgen/assets/44675551/6f3c0df7-e3d1-4a37-a617-8a35ade2d72d
https://github.com/gsgen3d/gsgen/assets/44675551/64e5662f-e8d5-4380-a2ac-540d6789f65b
https://github.com/gsgen3d/gsgen/assets/44675551/c9c25857-3b5c-4338-adb0-e8e0910c8260
https://github.com/gsgen3d/gsgen/assets/44675551/25e3a94e-5a3b-4e14-bdd5-bfbf44fc2b82
pip install -r requirements.txt
cd gs
./build.sh
python main.py --config-name=base prompt.prompt="<prompt>"
You can specify a different prompt for Point-E:
python main.py --config-name=base prompt.prompt="<prompt>" init.prompt="<point-e prompt>"
We support splat viewer now ! Click the captions of text-to-3D results in our project page to watch the assets in a WebGL based viwer. Example: a pineapple. This great viewer achieves > 40 FPS on my MacBook with M1 pro chip.
Start the Viewer by:
python vis.py <path-to-ckpt> --port <port>
If you are training on servers, tunneling the port using SSH
ssh -L <your_local_port>:<your_server_ip>:<your_server_port> <your_username>@<your_server>
then open the viewer in your host computer on port <your_local_port>
.
First set the PYTHONPATH
env var:
export PYTHONPATH="."
.ply
filepython utils/export.py <your_ckpt> --type ply
.splat
filepython utils/export.py <your_ckpt> --type splat
python utils/export.py <your_ckpt> --type mesh --batch_size 65536 --reso 256 --K 200 --thresh 0.1
where the 0|213630|2023-10-11|a_high_quality_photo_of_a_corgi
). The exported files are reside in the exports/<export-type>
.
-m
will work:python -m utils.export <your_ckpt> --type <export_type>
.ply
and .splat
files. Mesh exporting are coming soon.init.type="shap_e"
This code base is built upon the following awesome open-source projects:
Thank the authors for their remarkable job !