Tracking integration for image-to-video

[gary149]

• [ ] Integration with Inference API. Select at least one of the following:
  • [ ] Added a transformers pipeline
  • [ ] Added to Community Inference API for 3rd party library
  • [ ] Added to Community Inference API for generic
• [ ] Added basic UI elements (icon, order specification, etc) https://github.com/huggingface/huggingface.js/pull/317
• [ ] Added a widget

Integration guide: https://hf.co/docs/hub/adding-a-task

For widget (from @gary149 )

add image-to-video to PipelineType

https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt

<svg xmlns="http://www.w3.org/2000/svg" width="12" height="12" fill="none"><path fill="#000" fill-rule="evenodd" d="M1.43 1.2h6.63a.7.7 0 0 1 .7.7v4.44a.79.79 0 0 0-.34-.08h-.36V1.9H1.43v3.55l.9-.9a.7.7 0 0 1 .99 0l1.74 1.74a.79.79 0 0 0-.51.47L2.82 5.04l-1.4 1.4v2.09H4.5v.7H1.43a.7.7 0 0 1-.7-.7V1.9a.7.7 0 0 1 .7-.7ZM6.8 5.94l.32.32H5.5l.32-.32a.7.7 0 0 1 .99 0ZM6.2 4.52a1.05 1.05 0 1 1-1.17-1.75A1.05 1.05 0 0 1 6.2 4.52Zm-.4-1.16a.35.35 0 1 0-.38.58.35.35 0 0 0 .39-.58Zm3.9 5.16 1.57-1.26v3.14L9.7 9.15V10a.79.79 0 0 1-.79.79H5.77a.79.79 0 0 1-.79-.79V7.65a.79.79 0 0 1 .79-.78H8.9a.79.79 0 0 1 .79.78v.87Zm-3.93-.87v2.36H8.9V7.65H5.77Z" clip-rule="evenodd"/></svg>

[osanseviero]

cc @apolinario work needs to be done first in community inference API for the widget