引入视觉注意机制的目标跟踪方法综述
第40卷第4期自动化学报Vol.40,No.4 2014年4月ACTA AUTOMATICA SINICA April,2014
引入视觉注意机制的目标跟踪方法综述
黎万义1王鹏1乔红2
摘要视觉跟踪在无人飞行器、移动机器人、智能监控等领域有着广泛的应用,但由于目标外观和环境的变化,以及背景干扰等因素的存在,使得复杂场景下的鲁棒实时的目标跟踪成为一项极具挑战性的任务.视觉注意是人类视觉信息处理过程中的一项重要的心理调节机制,在视觉注意的引导下,人类能够从众多的视觉信息中快速地选择那些最重要、最有用、与当前行为最相关的感兴趣的视觉信息,特别地,人类能够快速指向感兴趣的目标,从而可以轻松地实现对目标的稳定跟踪.因此,将视觉注意机制引入到复杂场景下的目标跟踪中,有利于实现更为稳定和接近于人类认知机制的视觉跟踪算法.本文旨在对引入了视觉注意机制的目标跟踪方法进行综述.首先,介绍了视觉注意的基本概念及其代表性的计算模型;其次,对视觉注意与跟踪的内在关系进行了阐述;然后,对引入视觉注意机制的目标跟踪方法进行归纳、总结和分类,对代表性的方法进行介绍和分析;最后,对该类方法的特点和优势进行了讨论,并对未来的研究趋势进行了展望.
关键词目标跟踪,视觉注意,显著性,选择性注意,视觉认知
引用格式黎万义,王鹏,乔红.引入视觉注意机制的目标跟踪方法综述.自动化学报,2014,40(4):561?576
DOI10.3724/SP.J.1004.2014.00561
A Survey of Visual Attention Based Methods for Object Tracking
LI Wan-Yi1WANG Peng1QIAO Hong2
Abstract Visual tracking has been widely used in numerous applications,such as unmanned aerial vehicles,mobile robots and intelligent visual surveillance.Robust and real-time object tracking in complex scenes is a challenge task. Di?culties in tracking objects can arise due to changing appearance patterns of both the object and the environment, as well as factors such as background interference.Visual attention is one of the key mechanisms of visual perception which directs the processing resources to the visual data of the potentially most relevant,specially directs our gaze rapidly towards objects of interest in our visual environment and as a result humans can easily achieve stable object tracking. Therefore introducing the visual attention mechanism to the object tracking in complex scenes,will facilitate the realization of stable and humanoid tracking algorithms.This paper aims to review the state-of-the-art of visual attention based methods for tracking.Firstly,we introduce the basic concepts of visual attention and its representative computational models.Secondly,the relationship between visual attention and tracking is described.Thirdly,the attention-based visual tracking algorithms are classi?ed into?ve categories and detailed descriptions of representative methods in each category are provided,and their pros and cons are examined.Finally,we highlight the advantages of attention-based tracking methods and provide insights for future.
Key words Object tracking,visual attention,saliency,selective attention,visual cognition
Citation Li Wan-Yi,Wang Peng,Qiao Hong.A survey of visual attention based methods for object tracking.Acta Automatica Sinica,2014,40(4):561?576
视觉跟踪是计算机视觉中的一项重要任务,是指对视频序列中的目标状态进行持续推断的过程[1],
收稿日期2013-04-26录用日期2013-09-22
Manuscript received April26,2013;accepted September22, 2013
国家自然科学基金(61210009,61100098,61379097)资助Supported by National Natural Science Foundation of China (61210009,61100098,61379097)
本文责任编委贾云得
Recommended by Associate Editor JIA Yun-De
1.中国科学院自动化研究所精密感知与控制研究中心北京100190
2.中国科学院自动化研究所复杂系统管理与控制国家重点实验室北京100190
1.Research Center of Precision Sensing and Control,Institute of Automation,Chinese Academy of Sciences,Beijing100190
2.State Key Laboratory of Management and Control for Com-plex Systems,Institute of Automation,Chinese Academy of Sci-ences,Beijing100190其任务在于通过在视频的每一帧中定位目标,以生成目标的运动轨迹,并在每一时刻提供完整的目标区域[2].视觉跟踪技术在军事和民用方面都有着十分广泛的应用[1?6],军事方面包括无人飞行器、精确制导、空中预警、战场监视等,民用方面包括移动机器人、智能视频监控、智能交通系统、人机交互、虚拟现实等.
学术界和工业界在目标跟踪方面开展了大量工作,并取得了一定的进展[1?5,7].然而由于目标及场景的动态变化、遮挡、相似物干扰或者光线剧烈变化等复杂外界环境的干扰、目标的突变运动等因素的存在,使得复杂环境下鲁棒实时的目标跟踪仍然是一个具有挑战性的问题.