-
Notifications
You must be signed in to change notification settings - Fork 8.4k
Add Wan-FUN Camera Control models and Add WanCameraImageToVideo node #8013
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Conversation
I uploaded two different versions of quantization models to the community. |
You wrote two hyperlinks tha points to the same model Is the other one this one ? There are also two other models that are the same but without the camera word (just 1.3B-Control and 14B-Control) |
Yep sorry, it should be https://huggingface.co/alibaba-pai/Wan2.1-Fun-V1.1-14B-Control-Camera and https://huggingface.co/alibaba-pai/Wan2.1-Fun-V1.1-1.3B-Control-Camera |
I don't think the camera_conditions should be of type LATENT because it's confusing, it should be a different type like WAN_CAMERA_EMBEDDING. Can you also make it an optional input like the start image? You should also include a node that can generate some basic camera_conditions in this pr. |
I have updated the code based on your suggestions. |
VideoX-Fun recently released two models for camera control, Wan2.1-Fun-V1.1-1.3B-Control-Camera and Wan2.1-Fun-V1.1-14B-Control-Camera.
To support these two models, I add a new node WanCameraImageToVideo, for customizing inputs for camera control models.
I also added a node VideoAcc_CameraTrajectoryRecam to calculate Plücker embeddings as the input of WanCameraImageToVideo node, in my customized node repo.
Here I attached the JSON workflow for your reference.
wan2.1_fun_Camera_Control.json