Abstract/Summary

An image may convey a thousand words, but a video, composed of hundreds or thousands of image frames, tells a more intricate story. Despite significant progress in multimodal large language models (MLLMs), generating extended videos remains a formidable challenge. As of this writing, OpenAI’s Sora [1], the current state-of-the-art system, is still limited to producing videos of up to one minute in length. This limitation stems from the complexity of long video generation, which requires more than generative AI techniques for approximating density functions. Critical elements, such as planning, narrative construction, and spatiotemporal continuity, pose significant challenges. Integrating generative AI with a divide-and-conquer approach could improve scalability for longer videos while offering greater control. In this survey, we examine the current landscape of long video generation, covering foundational techniques such as GANs and diffusion models, video generation strategies, large-scale training datasets, quality metrics for evaluating long videos, and future research areas to address the limitations of existing video generation capabilities. We believe it would serve as a comprehensive foundation, offering extensive information to guide future advancements and research in the field of long video generation.