动作识别中局部时空特征的运动表示方法研究

近年来,基于局部时空特征的运动表征方法已被越来越多地运用于视频中的动作识别问题,相关研究人员已经提出了多种特征检测和描述方法,并取得了良好的效果。但上述方法在适应摄像头移动、光照以及穿着变化等方面还存在明显不足。为此,提出了基于时空兴趣点局部时空特征的运动表示方法,实现了基于时空单词的动作识别。首先采用基于Gabor滤波器和Gaussian滤波器相结合的检测算法从视频中提取时空兴趣点,然后抽取兴趣点的静态特征、运动特征和时空特征,并分别对运动进行表征,最后利用基于时空码本的动作分类器对动作进行分类识别。在Weizmann和KTH两个行为数据集进行了测试,实验结果表明：基于时空特征的运动表示能够更好地适应摄像头移动、光照变化以及施动者的穿着和动作差异等环境因素的影响,取得更好的识别效果。     

Modelling salient visual dynamics in videos

Automatic video annotation is a critical step for content-based video retrieval and browsing. Detecting the focus of interest in video frames automatically can benefit the tedious manual labeling process. However, producing an appropriate extent of visually salient regions in video sequences is a challenging task. Therefore, in this work, we propose a novel approach for modeling dynamic visual attention based on spatiotemporal analysis. Our model first detects salient points in three-dimensional video volumes, and then uses the points as seeds to search the extent of salient regions in a novel motion attention map. To determine the extent of attended regions, we use the maximum entropy in the spatial domain to analyze the dynamics derived by spatiotemporal analysis. Our experiment results show that the proposed dynamic visual attention model achieves high precision value of 70% and reveals its robustness in successive video volumes.     

基于局部时空特征的人体行为软分类识别

针对全局运动特征难以准确提取的问题,本文采用局部时空特征对人体行为进行表征。针对传统词袋中硬分类的方法量化误差大的不足,本文借鉴模糊聚类的思想,提出软分类的方法。根据兴趣点检测算法从视频中提取出视觉词汇,用K means算法对其进行聚类,建立码本。在计算分类特征时,首先计算待分类视觉词汇到码本中各个码字的距离,根据距离计算这个视觉词汇隶属于各个码字的概率,最后统计得到每个视频中各码字出现的频率。在Weizmann和KTH数据库对本文提出的人体行为识别算法进行验证,Weizmann库的识别率比传统的词袋算法提高8%,KTH库的识别率比传统的词袋算法提高9%,因此本文提出的算法能更有效地对人体行为进行识别。     

基于曲线拟合的视频稳像方法

根据人眼的视觉特性,提出一种基于曲线拟合的视频稳像方法。使用图像背景特征点对摄像机的全局运动进行估计,利用曲线拟合的方法计算出摄像机的抖动分量,并将曲线拟合的结果作为摄像机的主观运动方向,对其摄像机的抖动运动分量进行补偿,使图像位移矢量达到最小,以有效减少运动补偿后引起的图像信息丢失。对抖动角度在20°内移动摄像机拍摄的视频进行稳像处理,实验结果表明,该方法稳像后视频的抖动角度小于2°,视频图像信息的损失小于5%,具有较好的稳像效果,并且在稳像后保证了视频帧内容的完整性。     

Space-time adaptation for patch-based image sequence restoration

We present a novel space-time patch-based method for image sequence restoration. We propose an adaptive statistical estimation framework based on the local analysis of the bias-variance trade-off. At each pixel, the space-time neighborhood is adapted to improve the performance of the proposed patch-based estimator. The proposed method is unsupervised and requires no motion estimation. Nevertheless, it can also be combined with motion estimation to cope with very large displacements due to camera motion. Experiments show that this method is able to drastically improve the quality of highly corrupted image sequences. Quantitative evaluations on standard artificially noise-corrupted image sequences demonstrate that our method outperforms other recent competitive methods. We also report convincing results on real noisy image sequences     

可变场所的异常行为识别方法

在视觉分析中,人的同一动作在不同场景下会有截然不同的理解.为了判断在不同场景中行为是否为异常,在监控系统中使用双层词包模型来解决这个问题.把视频信息放在第1层包中,把场景动作文本词放在第2层包中.视频由一系列时空兴趣点组成的时空词典表示,动作性质由在指定场景下的动作文本词集合来确定.使用潜在语义分析概率模型(pLSA)不但能自动学习时空词的概率分布,找到与之对应的动作类别,也能在监督情况下学习在规定场景下运动文本词概率分布并区分出对应异常或正常行动结果.经过训练学习后,该算法可以识别新视频在相应场景下行为的异常或正常.     

结合全局与局部视频表示的视频异常检测算法

针对视频异常行为检测问题,提出结合全局与局部视频表示的视频异常检测算法.首先将输入视频连续多帧划分为视频块.再按空间位置将视频块划分为互不重叠的时空立方体,利用时空立方体运动特征构建基于空间位置的全局时空网格位置支持向量数据描述模型(SVDD).然后针对视频运动目标,提取局部纹理运动特征,采用SVDD获得围绕目标特征的超球体边界,构建运动目标正常行为模型.最后组合两部分以实现更全面的检测.公共数据集上的实验验证文中算法的有效性.     

动态背景下空时特性均显著的运动目标检测

从序列图像中提取变化区域是运动检测的主要作用,动态背景的干扰严重影响检测结果,使得有效性运动检测成为一项困难工作。受静态图像显著性检测启发,提出了一种新的运动目标检测方法,采用自底向上与自顶向下的视觉计算模型相结合的方式获取图像的空时显著性：先检测出视频序列中的空间显著性,在其基础上加入时间维度,利用改进的三帧差分算法获取具有运动目标的时间显著性,将显著性目标的检测视角由静态图像转换为空时性均显著的运动目标。实验和分析结果表明：新方法在摄像机晃动等动态背景中能较准确检测出空时均显著的运动目标,具有较高的鲁棒性。     

Fast and accurate global motion compensation

Video understanding has attracted significant research attention in recent years, motivated by interest in video surveillance, rich media retrieval and vision-based gesture interfaces. Typical methods focus on analyzing both the appearance and motion of objects in video. However, the apparent motion induced by a moving camera can dominate the observed motion, requiring sophisticated methods for compensating for camera motion without a priori knowledge of scene characteristics. This paper introduces two new methods for global motion compensation that are both significantly faster and more accurate than state of the art approaches. The first employs RANSAC to robustly estimate global scene motion even when the scene contains significant object motion. Unlike typical RANSAC-based motion estimation work, we apply RANSAC not to the motion of tracked features but rather to a number of segments of image projections. The key insight of the second method involves reliably classifying salient points into foreground and background, based upon the entropy of a motion inconsistency measure. Extensive experiments on established datasets demonstrate that the second approach is able to remove camera-based observed motion almost completely while still preserving foreground motion.     

Passive ranging using a moving camera

This article addresses the problem of determining the 3-dimensional locations of salient points in the environment of a moving camera based on a monocular image sequence obtained by the camera. The camera's translational and rotational velocities are assumed to be known approximately via inertial sensors. The motion of the camera is represented by a constant velocity model. Salient points in the image sequence are extracted using Gabor wavelets and tracked using labeled graph matching. The 3-D positions of the selected environmental points relative to the camera are then estimated recursively using an extended Kalman filter (EKF), after initialization by two-frame motion stereo. The motion parameters of the camera are also refined simultaneously. Experimental results on real data are given. © 1992 John Wiley & Sons, Inc.