[ICLR26] Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling

Paper PDF Project Page [Haoyu Wu](https://cintellifusion.github.io/)$^{1*}$, [Diankun Wu](https://github.com/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)
## 🎯 Overview ![](assets/main.png) **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. ## 🚀 News - [2026/01/26] Our Paper is accepted to [ICLR 2026](https://iclr.cc/) ! - [2025/10/8] We release the evaluation code for reprojection error and revisit error. - [2025/9/24] We release code and checkpoint. - [2025/9/22] [Geometry Forcing](https://geometryforcing.github.io/) is accepted to [NeurIPS 2025 NextVid Workshop](https://what-makes-good-video.github.io/) as an Oral! - [2025/7/10] We release the paper and the project. ## 💪 Get Started ### Setup Environments ```shell conda create -n geometryforcing python=3.10 -y conda activate geometryforcing pip install -r requirements.txt ``` ### Connect to Weights & Biases: We use Weights & Biases for logging. [Sign up](https://wandb.ai/login?signup=true) if you don't have an account, and *modify `wandb.entity` in `config.yaml` to your user/organization name*. ### Download Checkpoints and Data 1. Download pretrained checkpiont using huggingface: ```shell bash scripts/hf_download_checkpoints.sh ``` 2. Download pretrained checkpiont using modelscope: ```shell bash scripts/ms_download_checkpoints.sh ``` 3. Download and process RealEstate10k dataset to `data/real-estate-10k` The structure of RealEstate10K is exactly the same with DFoT. Please download RealEstate10k from dataset of DFoT from here [huggingface dataset](https://huggingface.co/kiwhansong/DFoT/tree/main/datasets). The structure should like this [wiki from DFoT](https://github.com/kwsong0113/diffusion-forcing-transformer/wiki/Dataset) ``` data/ ├── {dataset_name}/ │ ├── training/ │ │ ├── video_xxx.mp4 │ │ ├── ... │ ├── validation/ │ │ ├── video_xxx.mp4 │ │ ├── ... │ ├── test/ │ │ ├── video_xxx.mp4 │ │ ├── ... │ ├── metadata/ │ │ ├── training.pt │ │ ├── validation.pt │ │ ├── test.pt ``` ### Generating Videos with Pretrained Models 1. Single Image to Long Video (256 Frames): ```shell bash scripts/eval_geometry_forcing.sh ``` 2. Single Image to Rotation Video (16 Frames): ```shell bash scripts/eval_geometry_forcing_rotation.sh ``` ### Training Geometry Forcing To train Geometry Forcing, run the following command: ```shell bash scripts/train_geometry_forcing.sh ``` ### Evaluation for Reprojection Error and Revisit Error To evaluate the reprojection error and revisit error, please follow the instructions in [README_EVAL.md](README_EVAL.md). ## 📜 Citation If you find our work useful for your research, please consider citing our paper: ``` @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} } ``` # Research template attribution This repo is forked from [Boyuan Chen](https://boyuan.space/)'s research template [repo](https://github.com/buoyancy99/research-template). By its license, you must keep the above sentence in `README.md` and the `LICENSE` file to credit the author.