search

Francis-Rings / MotionEditor

[CVPR2024] MotionEditor is the first diffusion-based model capable of video motion editing.
113 stars 6 forks source link

Dataset #5

Open AsukaCamellia opened 1 month ago

AsukaCamellia commented 1 month ago

Thanks for your wonderful work. Where is the preprocessing of the dataset?

Francis-Rings commented 1 month ago

Thanks for your attention! We implement SAM and OpenPose to exact protagonist masks and pose skeletons of the given videos, respectively.

bing32475 commented 3 weeks ago

Thank you for your excellent work. Where can I download the dataset used in the project?

Francis-Rings commented 1 week ago

Thanks for your attention! The codes of data preparation are available to the public. MotionEditor is a one-shot learning-based diffusion model. Consequently, MotionEditor doesn't need any training datasets. If you would like to access the source videos utilized for inference, please feel free to send me an email for further details.