Variable length frame conditioning for infinite length video generation. Can be used to generate videos from text that continue from an existing video or image. It can also generate the initial image using a standard text-to-image model from huggingface. Developed in collaboration with motexture and based on vseq2vseq.
Repo for inference can be found at ContinuitySeq.
- Downloads last month
- 3
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 pipeline_tag.