This model performs image-to-video generation based on the paper FlexWorld: Progressively Expanding 3D Scenes for Flexible-View Synthesis.
Project page: https://ml-gsai.github.io/FlexWorld
Code: https://github.com/ml-gsai/FlexWorld
Usage Example
A basic example of generating a static scene video given an image and a camera trajectory:
# You can utilize our CamPlanner class to freely construct the desired trajectory at line 13 in the `video_generate.py` file.
python video_generate.py --input_image_path ./assets/room.png --output_dir ./results-single-traj
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.