Skip to content

Latest commit

 

History

History
167 lines (144 loc) · 8.96 KB

README.md

File metadata and controls

167 lines (144 loc) · 8.96 KB
Raw

ControlVideo

Official pytorch implementation of "ControlVideo: Training-free Controllable Text-to-Video Generation"

arXiv Project HuggingFace demo Replicate visitors


ControlVideo adapts ControlNet to the video counterpart without any finetuning, aiming to directly inherit its high-quality and consistent generation

News

Setup

1. Download Weights

All pre-trained weights are downloaded to checkpoints/ directory, including the pre-trained weights of Stable Diffusion v1.5, ControlNet 1.0 conditioned on canny edges, depth maps, human poses, and ControlNet 1.1 in here. The flownet.pkl is the weights of RIFE. The final file tree likes:

checkpoints
├── stable-diffusion-v1-5
├── sd-controlnet-canny
├── sd-controlnet-depth
├── sd-controlnet-openpose
├── ...
├── flownet.pkl

2. Requirements

conda create -n controlvideo python=3.10
conda activate controlvideo
pip install -r requirements.txt

Note: xformers is recommended to save memory and running time. controlnet-aux is updated to version 0.0.6.

Inference

To perform text-to-video generation, just run this command in inference.sh:

python inference.py \
    --prompt "A striking mallard floats effortlessly on the sparkling pond." \
    --condition "depth" \
    --video_path "data/mallard-water.mp4" \
    --output_path "outputs/" \
    --video_length 15 \
    --smoother_steps 19 20 \
    --width 512 \
    --height 512 \
    --frame_rate 2 \
    --version v10 \
    # --is_long_video

where --video_length is the length of synthesized video, --condition represents the type of structure sequence, --smoother_steps determines at which timesteps to perform smoothing, --version selects the version of ControlNet (e.g., v10 or v11), and --is_long_video denotes whether to enable efficient long-video synthesis.

Visualizations

ControlVideo on depth maps

"A charming flamingo gracefully wanders in the calm and serene water, its delicate neck curving into an elegant shape." "A striking mallard floats effortlessly on the sparkling pond." "A gigantic yellow jeep slowly turns on a wide, smooth road in the city."
"A sleek boat glides effortlessly through the shimmering river, van gogh style." "A majestic sailing boat cruises along the vast, azure sea." "A contented cow ambles across the dewy, verdant pasture."

ControlVideo on canny edges

"A young man riding a sleek, black motorbike through the winding mountain roads." "A white swan movingon the lake, cartoon style." "A dusty old jeep was making its way down the winding forest road, creaking and groaning with each bump and turn."
"A shiny red jeep smoothly turns on a narrow, winding road in the mountains." "A majestic camel gracefully strides across the scorching desert sands." "A fit man is leisurely hiking through a lush and verdant forest."

ControlVideo on human poses

"James bond moonwalk on the beach, animation style." "Goku in a mountain range, surreal style." "Hulk is jumping on the street, cartoon style." "A robot dances on a road, animation style."

Long video generation

"A steamship on the ocean, at sunset, sketch style." "Hulk is dancing on the beach, cartoon style."

Citation

If you make use of our work, please cite our paper.

@article{zhang2023controlvideo,
  title={ControlVideo: Training-free Controllable Text-to-Video Generation},
  author={Zhang, Yabo and Wei, Yuxiang and Jiang, Dongsheng and Zhang, Xiaopeng and Zuo, Wangmeng and Tian, Qi},
  journal={arXiv preprint arXiv:2305.13077},
  year={2023}
}

Acknowledgement

This work repository borrows heavily from Diffusers, ControlNet, Tune-A-Video, and RIFE. The code of HuggingFace demo borrows from fffiloni/ControlVideo. Thanks for their contributions!

There are also many interesting works on video generation: Tune-A-Video, Text2Video-Zero, Follow-Your-Pose, Control-A-Video, et al.