Veo


Veo, a video generation model, uses advanced understanding of natural language and visual semantics to create videos that accurately capture the nuance and tone of a prompt. The model can generate videos from text prompts, editing commands, or reference images, and can also extend video clips beyond 60 seconds while maintaining visual consistency. [summary] [comments]


这是一个从 https://deepmind.google/technologies/veo/ 下的原始话题分离的讨论话题