(SVD XT 1.1) Image-to-Video is a latent diffusion model released this month that is trained to generate short video clips from an image conditioning.
This model was trained to generate 25 frames at resolution 1024x576 given a context frame of the same size, finetuned from SVD Image-to-Video [25 frames].
Fine tuning was performed with fixed conditioning at 6FPS and Motion Bucket Id 127 to improve the consistency of outputs without the need to adjust hyper parameters. These conditions are still adjustable and have not been removed. Performance outside of the fixed conditioning settings may vary compared to SVD 1.0.
---
This model is intended for research purposes only.
While this model is open source, but please respect the hard work of the people at Stability AI and review the acceptable use cases set forth by the team.
Developed by: Stability AI
Funded by: Stability AI
Model type: Generative image-to-video model
Finetuned from model: SVD Image-to-Video [25 frames]