arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.16407cs.CV
收藏

长视频生成综述:挑战、方法与前景

A Survey on Long Video Generation: Challenges, Methods, and Prospects

Xidian University(西安电子科技大学);The University of Sydney(悉尼大学);Shanghai AI Laboratory(上海人工智能实验室);Shanghai Jiao Tong University(上海交通大学)

Chengxuan Li, Di Huang, Zeyu Lu, Yang Xiao, Qingqi Pei, Lei Bai

AI总结:

本文首次对长视频生成的最新进展进行综述,总结分治法与时间自回归两大范式,探讨常用模型、数据集与评估指标,并指出新兴挑战与未来方向,为该领域提供重要参考。

摘要:

视频生成是一个快速发展的研究领域,因其广泛的应用而备受关注。该领域的一个关键方面是长视频生成,它带来了独特的挑战与机遇。本文首次对长视频生成的最新进展进行了综述,并将其总结为两个关键范式:分治法与时间自回归。我们深入探讨了每种范式中常用的模型,包括网络设计和条件技术等方面。此外,我们对数据集和评估指标进行了全面的概述与分类,这些对于推进长视频生成研究至关重要。在总结现有研究的同时,我们还讨论了这一动态领域的新兴挑战与未来方向。我们希望本综述能成为长视频生成领域研究人员和从业者的重要参考。

英文摘要:

Video generation is a rapidly advancing research area, garnering significant attention due to its broad range of applications. One critical aspect of this field is the generation of long-duration videos, which presents unique challenges and opportunities. This paper presents the first survey of recent advancements in long video generation and summarises them into two key paradigms: divide and conquer temporal autoregressive. We delve into the common models employed in each paradigm, including aspects of network design and conditioning techniques. Furthermore, we offer a comprehensive overview and classification of the datasets and evaluation metrics which are crucial for advancing long video generation research. Concluding with a summary of existing studies, we also discuss the emerging challenges and future directions in this dynamic field. We hope that this survey will serve as an essential reference for researchers and practitioners in the realm of long video generation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑