Image to Video
Hey,
thanks i realy enjoy the model.
Any news/release-date about the Image2Video model?
Cant wait to test that :)
Find anything?
Chien cheval
How to access it
Close to two months, they are really quiet. Surprisingly the paper mentions about Image to Video
The open source community is eagerly awaiting the Hunyuan image-to-video model. Native text-to-video in ComfyUI has been integrated for awhile now, so that box can be checkmarked ✅: https://comfyanonymous.github.io/ComfyUI_examples/hunyuan_video/
In the meantime leapfusion has done impressive work trying to fill the gap with an image-to-video lora that can be used with the hunyuan text-to-video model:
https://github.com/AeroScripts/leapfusion-hunyuan-image2video
A dedicated image-to-video model would really be a blessing to the open-source-community.
is there a non-comfyui way of doing this with just scripting?
Yes, though I haven't tried it. There are directions in the leapfusion repo on how to run their lora without comfyui. Git clone the repo, download the models, set up the environment (probably the hardest part), then just have to input some command-line args: https://github.com/AeroScripts/leapfusion-hunyuan-image2video