Control-DINO: Feature Space Conditioning for Controllable Image-to-Video Diffusion

git clone --recursive https://github.com/d3ixi/ControlDINO.git
cd ControlDINO

conda create -n ControlDINO python=3.10 -y
conda activate ControlDINO
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu129
pip install -r requirements.txt

python download.py 

conda install -c conda-forge git-lfs
git lfs install
git lfs pull

./scripts/inference_recon.sh    # reconstruction 
./scripts/inference_style.sh    # styling
./scripts/inference_vkitti.sh   # vkitti

python app.py

@article{dominici2026controldinofeaturespaceconditioning,
      title={Control-DINO: Feature Space Conditioning for Controllable Image-to-Video Diffusion}, 
      author={Edoardo A. Dominici and Thomas Deixelberger and Konstantinos Vardis and Markus Steinberger},
      year={2026},
      eprint={2604.01761},
      archivePrefix={arXiv},
      url={https://arxiv.org/abs/2604.01761}, 
}