ShotPlan: Cinematic Video Generation with Learnable Planning Token
For video generation researchers, this provides a novel framework for coherent multi-shot narrative generation, addressing a key limitation of current single-shot models.
ShotPlan introduces learnable planning tokens with FRoPE for explicit multi-shot cinematic video generation, significantly outperforming existing methods in shot management and inter-shot consistency.
Current video generation models achieve impressive results in single-shot generation, yet remain limited in cinematic video generation, where coherent narratives and effective multi-shot composition require explicit shot planning. To address this challenge, we propose ShotPlan, a framework for explicit multi-shot cinematic video generation built upon a video diffusion foundation model. Our method introduces learnable planning tokens that capture shot-level transition cues and can be seamlessly integrated with the original video generation tokens to control transition timestamps. Unlike standard video generation tokens, the proposed planning tokens are equipped with Fractional Temporal Rotary Position Embedding (FRoPE), enabling shot transitions to be modeled at the frame level. Experiments demonstrate that ShotPlan significantly outperforms existing cinematic video generation methods, offering more flexible shot management and stronger inter-shot consistency.