BindWeave: Subject-Consistent Video Generation via Cross-Modal Integration
This addresses the challenge of generating videos with consistent subjects and complex interactions for applications in content creation and simulation, representing a strong incremental advance over existing models.
The paper tackles the problem of subject-consistent video generation by proposing BindWeave, a framework that integrates a multimodal large language model with a diffusion transformer to ground entities and disentangle interactions from prompts, achieving superior performance on the OpenS2V benchmark in subject consistency, naturalness, and text relevance.
Diffusion Transformer has shown remarkable abilities in generating high-fidelity videos, delivering visually coherent frames and rich details over extended durations. However, existing video generation models still fall short in subject-consistent video generation due to an inherent difficulty in parsing prompts that specify complex spatial relationships, temporal logic, and interactions among multiple subjects. To address this issue, we propose BindWeave, a unified framework that handles a broad range of subject-to-video scenarios from single-subject cases to complex multi-subject scenes with heterogeneous entities. To bind complex prompt semantics to concrete visual subjects, we introduce an MLLM-DiT framework in which a pretrained multimodal large language model performs deep cross-modal reasoning to ground entities and disentangle roles, attributes, and interactions, yielding subject-aware hidden states that condition the diffusion transformer for high-fidelity subject-consistent video generation. Experiments on the OpenS2V benchmark demonstrate that our method achieves superior performance across subject consistency, naturalness, and text relevance in generated videos, outperforming existing open-source and commercial models.