Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling
Paper
•
2507.07982
•
Published
•
33
Haoyu Wu$^{1*}$, Diankun Wu $^{2*}$, Tianyu He $^{1†}$, Junliang Guo $^{1}$, Yang Ye $^{1}$, Yueqi Duan $^{2}$, Jiang Bian $^{1}$
$^1$ Microsoft Research $^2$ Tsinghua University
($^*$ Equal Contribution. † Project Lead)
@article{wu2025geometryforcing,
title={Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling},
author={Wu, Haoyu and Wu, Diankun and He, Tianyu and Guo, Junliang and Ye, Yang and Duan, Yueqi and Bian, Jiang},
journal={arXiv preprint arXiv:2507.07982},
year={2025}
}
Geometry Forcing (GF) Overview.
(a) Our proposed GF paradigm enhances video diffusion models by aligning with geometric features from VGGT.
(b) Compared to DFoT, our method generates more temporally and geometrically consistent videos.
(c) While baseline features fail to reconstruct meaningful 3D geometry, GF-learned features enable accurate 3D reconstruction.
conda create -n geometryforcing python=3.10 -y
conda activate geometryforcing
pip install -r requirements.txt
We use Weights & Biases for logging. Sign up if you don't have an account, and modify wandb.entity in config.yaml to your user/organization name.
bash scripts/hf_download_checkpoints.sh
bash scripts/ms_download_checkpoints.sh
data/real-estate-10kbash scripts/eval_geometry_forcing.sh
bash scripts/eval_geometry_forcing_rotation.sh
bash scripts/train_geometry_forcing.sh