Multi-shot video training data for ShotPlan: Cinematic Video Generation with Learnable Planning Token.
š» Code:
https://github.com/Pensioner-11/ShotPlan
š¤ Models: ShotPlan-Wan2.1-T2V-14B Ā· ShotPlan-Wan2.2-T2V-A14B-HighNoise
Each sample is an 80-frame (5 s @⦠See the full description on the dataset page:
https://huggingface.co/datasets/Pensioner/shotplan.