how to finetune with image to video task？

[trouble-maker007]

I test the image to video and compare with the dynamic crafter result, EasyAnimateV3 is not good as it, and I want to finetune on my own data
Is it inpaint mode is to train image to video task?

[yunkchen]

https://github.com/aigc-apps/EasyAnimate/blob/main/scripts/train_lora.sh

[trouble-maker007]

@yunkchen thanks for the quick response, does https://github.com/aigc-apps/EasyAnimate/blob/main/scripts/train.sh with can also finetune with image to video task, but need more gpus?

[yunkchen]

@yunkchen thanks for the quick response, does https://github.com/aigc-apps/EasyAnimate/blob/main/scripts/train.sh with can also finetune with image to video task, but need more gpus?

you are right.