深度光流估计方法研究进展  

Research Progress of Deep Optical Flow Estimation

作　　者：祖曰然 包秀国[2] 唐文忠[1] 高科[3] 张冬明 Zu Yueran;Bao Xiuguo;Tang Wenzhong;Gao Ke;Zhang Dongming(School of Computer Science and Technology,Beihang University,Beijing 100191;The National Computer Network Emergency Response Technical Team Coordination Center of China,Beijing 100029;Key Laboratory of Intelligent Information Processing,Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190)

机构地区：[1]北京航空航天大学计算机学院,北京100191 [2]国家计算机网络与信息安全管理中心,北京100029 [3]中国科学院计算技术研究所智能信息处理重点实验室,北京100190

出　　处：《计算机辅助设计与图形学学报》2021年第2期310-320,共11页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(51475025);国家重点研发计划(2016YFB0801203);北京市科学技术委员会基金(Z171100000117010).

摘　　要：结合深度学习模型实现光流端到端的计算是当前计算机视觉领域的一个研究热点.文中对基于深度学习的光流估计方法进行总结和梳理.首先,介绍了光流的起源与定义;其次,总结了现有的数据集合和评价指标;最重要的是,着重从3个方面回顾了深度光流估计方法,包括有监督的深度光流估计方法、无监督的深度光流估计方法以及对现有光流估计方法的性能对比分析.分析表明,参照传统方法设计小而轻且泛化性能好的深度光流网络是未来的研究方向.在此基础上,进一步分析和介绍了当下光流估计与视频分析任务联合学习的一系列代表性方法,指出了设计由任务驱动的深度光流网络是很有应用价值的研究方向.最后,总结了深度光流估计存在的问题和挑战,并对未来工作进行展望.Combining the deep learning model to compute end-to-end optical flow is a hot topic in current computer vision field.The optical flow estimation methods based on deep learning are summarized and reviewed.Firstly,the origin and concept of optical flow is introduced.Secondly,the optical flow datasets and evaluation metrics are summarized.Most importantly,classical methods are introduced and the deep optical flow estimation methods are reviewed in three aspects,including supervised deep optical flow estimation methods,unsupervised deep optical flow estimation methods and the performance of these methods.The analysis shows designing compact and generalized deep optical flow model is the future research direction.On this basis,the joint learning of optical flow estimation and specific video analysis tasks are introduced.It is pointed out that the design of task-driven deep optical flow network is a valuable research direction in practice application.Finally,the problems and challenges of deep optical flow estimation are summarized and the future work of this field is prospected.

关 键 词：光流 光流估计 深度学习 视频分析 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]