search

jjihwan / FIFO-Diffusion_public

Official implementation of FIFO-Diffusion: Generating Infinite Videos from Text without Training (NeurIPS 2024)
https://jjihwan.github.io
397 stars 26 forks source link

Support for Dyanmicraft #18

Closed JACKHAHA363 closed 5 months ago

JACKHAHA363 commented 5 months ago

Dynamicraft is the img2video version of VideoCraft. How easy is it to extend current VideoCraft implementation?

jjihwan commented 5 months ago

FIFO-Diffusion is designed for T2V models, so I cannot assume how difficult it is to implement an I2V version of FIFO at this time. However, since DynamiCrafter is a model where the input image is not the first frame of the generated video, contrary to SV3D or other general I2V models, I believe it is not compatible with our method.