search

Vchitect / Latte

Latte: Latent Diffusion Transformer for Video Generation.
Apache License 2.0
1.45k stars 147 forks source link

run bash sample/t2v.sh,but why? #24

Open liwei0826 opened 4 months ago

liwei0826 commented 4 months ago

try generating videos from text，why? image

https://github.com/Vchitect/Latte/assets/48194881/402b169c-b78d-4c7a-ac27-a3ea80b90fe5

maxin-cn commented 4 months ago

try generating videos from text，why? image

t2vA_dog_in_astronaut_suit_and_sunglasses_floating_in_space._0000webv-imageio.mp4

Hi, I just tested my code and the t2v model did not produce the quality video as you showed. Could you please check your code or pull down the code again? Thank you for your interest.

https://github.com/Vchitect/Latte/assets/38418898/3102b7c8-8beb-41cb-a582-a8f0caff08ee