Closed aejion closed 3 weeks ago
Hi, @aejion, thanks for your interest in our project. SVD can predict a continuous video from a given image, thus ensuring better consistency between different generated viewpoints. This is a capability that image diffusion models lack.
Thanks for your quick reply!
Hi! Great work!
I would like to know if it is possible for NVS-Solver to use an image diffusion model for interpolation. Additionally, could you explain why you chose to use a video diffusion model for interpolation?