LoopGaussian: Creating 3D Cinemagraph with Multi-view Images via Eulerian Motion Field

Jiyang Li^1*, Lechao Cheng^2*, Zhangye Wang¹, Tingting Mu³, Jingxuan He^2✉
*Equal contribution. ^✉Corresponding author.
¹Zhejiang University, ²Hefei University of Technology, ³The University of Manchester

Environment

• CUDA = 12.1
• Pytorch = 2.1.1
• Python = 3.11

pip install -r requirements.txt

Train

Train 3D-GS

python gaussian_splatting/train.py -s {images_path}

Train LoopGaussian

python train_loop.py --task_name ${task_name}

The rendering result will be saved in output/{task_name}/visual.mp4