🎬 DomainShuttle — Subject-driven Text-to-Video

Freeform open-domain subject-driven video generation. Give one to three reference images of subjects (people, objects, characters) and a text prompt; the model re-renders those subjects into a new video — including cross-domain restyling (e.g. real photo → Pixar style).

Built on a two-expert (high/low noise) MoE finetune of Wan2.2-A14B. Weights: CNcreator0331/DomainShuttle_weight · Code: HKUST-C4G/DomainShuttle · Project page