一种改进的窄带算法在彩色序列图像人脸跟踪中的应用

该文提出了一种基于几何主动轮廓模型的人脸跟踪方法.通过直方图反向投影,使人脸区域表现为一个一致性区域与背景相区别.研究了一种改进的窄带算法实现曲线演化:以等间隔分布的节点表示运动曲线,只在这些节点上计算Level set函数的变化值,窄带区内其余点的Level set值的更新通过插值和查表的方法实现;根据节点的局部图像信息决定节点的运动方向和时间步长值.实验表明该算法能在满足一定精度的前提下,快速地对运动人脸进行跟踪.     

一种基于邻域支持准则的双层立体匹配算法

本文给出一种基于邻域支持准则的双层立体匹配算法，该算法以LOG边缘检测器输出的边缘点为匹配基元，利用图象相关特性和匹配相容性约束构成两层匹配系统．第一层以反映图象灰度统计特性的局部相关约束为准则．第二层以第一层输出为引导，从图象结构和视差相容方面寻求全局意义上的最佳匹配，从而使系统对复杂场景有较好的自适应性．实验结果表明，该算法结构简单，稳定可靠，具有处理复杂场景的能力．     

Tracking a dynamic set of feature points

We address the problems of tracking a set of feature points over a long sequence of monocular images as well as how to include and track new feature points detected in successive frames. Due to the 3-D movement of the camera, different parts of the images exhibit different image motion. Tracking discrete features can therefore be decomposed into several independent and local problems. Accordingly, we propose a localized feature tracking algorithm. The trajectory of each feature point is described by a 2-D kinematic model. Then to track a feature point, an interframe motion estimation scheme is designed to obtain the estimates of interframe motion parameters. Subsequently, using the estimates of motion parameters, corresponding points are identified to subpixel accuracy. Afterwards, the temporal information is processed to facilitate the tracking scheme. Since different feature points are tracked independently, the algorithm is able to handle the image motion arising from general 3-D camera movements. On the other hand, in addition to tracking feature points detected at the beginning, an efficient way to dynamically include new points extracted in subsequent frames is devised so that the information in a sequence is preserved. Experimental results for several image sequences are also reported.     

基于多尺度细节增强的面部表情识别方法

人类面部表情是其心理情绪变化的最直观刻画,不同人的面部表情具有很大差异,现有表情识别方法均利用面部统计特征区分不同表情,其缺乏对于人脸细节信息的深度挖掘。根据心理学家对面部行为编码的定义可以看出,人脸的局部细节信息决定了其表情意义。因此该文提出一种基于多尺度细节增强的面部表情识别方法,针对面部表情受图像细节影响较大的特点,提出利用高斯金字塔提取图像细节信息,并对图像进行细节增强,从而强化人脸表情信息。针对面部表情的局部性特点,提出利用层次结构的局部梯度特征计算方法,描述面部特征点局部形状特征。最后,使用支持向量机(SVM)对面部表情进行分类。该文在CK+表情数据库中的实验结果表明,该方法不仅验证了图像细节对面部表情识别过程的重要作用,而且在小规模训练数据下也能够得到非常好的识别结果,表情平均识别率达到98.19%。     

Super-resolution without dense flow

Super-resolution is a widely applied technique that improves the resolution of input images by software methods. Most conventional reconstruction-based super-resolution algorithms assume accurate dense optical flow fields between the input frames, and their performance degrades rapidly when the motion estimation result is not accurate enough. However, optical flow estimation is usually difficult, particularly when complicated motion is presented in real-world videos. In this paper, we explore a new way to solve this problem by using sparse feature point correspondences between the input images. The feature point correspondences, which are obtained by matching a set of feature points, are usually precise and much more robust than dense optical flow fields. This is because the feature points represent well-selected significant locations in the image, and performing matching on the feature point set is usually very accurate. In order to utilize the sparse correspondences in conventional super-resolution, we extract an adaptive support region with a reliable local flow field from each corresponding feature point pair. The normalized prior is also proposed to increase the visual consistency of the reconstructed result. Extensive experiments on real data were carried out, and results show that the proposed algorithm produces high-resolution images with better quality, particularly in the presence of large-scale or complicated motion fields.     

Facial model adaptation from a monocular image sequence using a textured polygonal model

Although several algorithms have been proposed for facial model adaptation from image sequences, the insufficient feature set to adapt a full facial model, imperfect matching of feature points, and imprecise head motion estimation may degrade the accuracy of model adaptation. In this paper, we propose to resolve these difficulties by integrating facial model adaptation, texture mapping, and head pose estimation as cooperative and complementary processes. By using an analysis-by-synthesis approach, salient facial feature points and head profiles are reliably tracked and extracted to form a growing and more complete feature set for model adaptation. A more robust head motion estimation is achieved with the assistance of the textured facial model. The proposed scheme is performed with image sequences acquired with single uncalibrated camera and requires only little manual adjustment in the initialization setup, which proves to be a feasible approach for facial model adaptation.     

Estimation of global motion parameters by complex linear regression

Global motion is very likely to occur in image sequences analysis. For example, it arises if the observer is moving during the sequence acquisition (ego-motion). Our aim is to get a simple method to estimate in a reliable may a set of parameters that can take into account the presence of a global motion component, using only local information. The novelty of our approach is in regarding spatial shift, change of scale, and rotation (corresponding to usual camera effects such as pan and zoom) as a two-dimensional (2-D) Doppler effect. The mathematical treatment is carried on in the complex plane, so that the results can be easily deduced as an extension of the one-dimensional (1-D) case; in this way, we obtain simple expressions, well suited for a practical realization of the estimate. The method has been experimentally validated by both real pictures with a synthetic motion and real image sequences.     

A hybrid model-based image coding system for very low bit-ratecoding

This paper proposes a two-stage global motion estimation method and a hybrid coding algorithm for the model-based coding. In the first stage, global motion is estimated by the feature-based algorithm. The estimated result is further refined by the gradient method. The two-stage estimation algorithm is performed hierarchically to remove the influence of the local facial expression. In the estimation process, we also utilize the steerable pyramid to improve accuracy. The facial expression region is coded by the clip-and-paste method, and is predicted by a classified prototype coding technique. The prototype coding technique can greatly improve the coding efficiency for the facial expression. The areas, which are difficult to be described by generic models, are encoded by the proposed hierarchical motion segmentation algorithm. We segment the image to different moving areas which can be modeled by affine models in a split-and-merge manner. The segmented results with the estimated affine models can give good prediction for the unmodeled regions. Computer simulation results show that the proposed complete model-based coding scheme gives very good performance in terms of peak signal-to-noise ratio and compression ratio     

Projection-based spatially adaptive reconstruction ofblock-transform compressed images

At the present time, block-transform coding is probably the most popular approach for image compression. For this approach, the compressed images are decoded using only the transmitted transform data. We formulate image decoding as an image recovery problem. According to this approach, the decoded image is reconstructed using not only the transmitted data but, in addition, the prior knowledge that images before compression do not display between-block discontinuities. A spatially adaptive image recovery algorithm is proposed based on the theory of projections onto convex sets. Apart from the data constraint set, this algorithm uses another new constraint set that enforces between-block smoothness. The novelty of this set is that it captures both the local statistical properties of the image and the human perceptual characteristics. A simplified spatially adaptive recovery algorithm is also proposed, and the analysis of its computational complexity is presented. Numerical experiments are shown that demonstrate that the proposed algorithms work better than both the JPEG deblocking recommendation and our previous projection-based image decoding approach.     

基于IEKF视觉运动分析递归算法的研究

本文提出基于图象序列上特征点的图象平面轨迹进行运动参数估计，以提取运动物体的姿态、速度以及外部环境内所感兴趣点的位置参数，由于图象特征点轨迹与待估计的参数有关，因此能够组合长序列图象内的信息并利用ＩＥＫＦ递归地估计未知的运动及结构参数。     

Joint modeling and reconstruction of a compressively-sensed set of correlated images

Employing correlation among images for improved reconstruction in compressive sensing is a conceptually attractive idea, although developing efficient modeling strategies and reconstruction algorithms are often the key to achieve any potential benefit. This paper presents a novel modeling strategy and an efficient reconstruction algorithm for processing a set of correlated images, jointly taking into consideration inter-image correlation, intra-image correlation and inter-channel correlation. The approach starts with joint modeling of the entire image set in the gradient domain, which supports simultaneous representation of local smoothness, nonlocal self-similarity of every single image, and inter-image correlation. Then an efficient algorithm is proposed to solve the joint formulation, using a Split-Bregman-based technique. Furthermore, to support color image reconstruction, the proposed algorithm is extended by using the concept of group sparsity to explore inter-channel correlation. The effectiveness of the proposed approach is demonstrated with extensive experiments on both grayscale and color image sets. Results are also compared with recently proposed compressive sensing recovery algorithms.     

基于非线性SIFT框架SAR图像匹配

针对基于尺度不变特征变换（SIFT）的图像匹配算法性能受到SAR图像中严重斑点噪声而性能降低的问题，提出了一种改进的非线性尺度构建的SIFT算法，主要改进在于：在尺度空间构建阶段，该算法通过将滚动引导滤波器嵌入到尺度空间构造的过程中来生成多尺度图像金字塔，在去除斑噪的同时并保持边缘的方面表现出了较其他尺度空间构建算法更好的效果；在特征检测阶段，提出了一种使用ROEWA算子和Harris-Laplace检测算子相结合的特征点检测算法，有效地抑制SAR图像中的虚假特征点，并准确地提取具有高位置精度和低误差率的不变特征点。3种不同类型的仿真和真实SAR图像对该算法进行了检验，并与其他2种基于SIFT的方法相比较，实验结果表明，该算法在匹配精度和内联点比率方面可以实现更好的性能。     

Content-based adaptive spatio-temporal methods for MPEG repair

Block loss and propagation error due to cell loss or missing packet information during the transmission over lossy networks can cause severe degradation of block and predictive-based video coding. Herein, new fast spatial and temporal methods are presented for block loss recovery. In the spatial algorithm, missing block recovery and edge extention are performed by pixel replacement based on range constraints imposed by surrounding neighborhood edge information and structure. In the temporal algorithm, an adaptive temporal correlation method is proposed for motion vector (MV) recovery. Parameters for the temporal correlation measurement are adaptively changed in accordance to surrounding edge information of a missing macroblock (MB). The temporal technique utilizes pixels in the reference frame as well as surrounding pixels of the lost block. Spatial motion compensation is applied after MV recovery when the reference frame does not have sufficient information for lost MB restoration. Simulations demonstrate that the proposed algorithms recover image information reliably using both spatial and temporal restoration. We compare the proposed algorithm with other procedures with consistently favorable results.     

基于可变区域拟合水平集算法的改进

基于可变区域拟合水平集算法利用图像的局部区域信息,在活动曲线演化控制参数的手工设置使其应用受到了限制。本文提出了将灰度信息图像匹配原理应用到RSF模型中,根据计算相邻演化图像的相关系数实现迭代的自适应停止。实验结果表明,改进的RSF模型克服了自动设置迭代次数的缺点,实现了迭代的自适应停止,而且对弱边缘不连续图像能够有效地实现,节省了时间,提高了分割效率。     

基于多特征融合词包模型的SAR目标鉴别算法

针对复杂场景中的SAR目标鉴别问题,该文提出一种基于多特征融合词包(Bag-of-Words, BoW)模型的SAR目标鉴别算法。在BoW模型底层特征提取阶段,算法采用SAR-SIFT特征描述局部区域的形状信息;同时,采用该文基于传统鉴别特征提出的一组新的SAR图像局部特征描述局部区域的对比度信息和纹理信息。对于BoW模型中多个底层特征的融合,算法采用图像层的特征融合方式生成图像的全局鉴别特征,其中各单底层特征BoW模型特征的权系数通过L2范数约束的多核学习方法训练得到。在MiniSAR实测SAR图像数据上的目标鉴别实验表明,与基于传统鉴别特征以及单底层特征BoW模型特征的鉴别算法相比较,该文基于多特征融合BoW模型SAR目标鉴别算法具有更好的鉴别性能。     

Predictive coding of images using an adaptive intra-frame predictor and motion compensation

In this article we present an image predictive coding method using both intra- and inter-frame predictors. The intra-frame predictor is an adaptive FIR filter using the well-known LMS algorithm to track continuously spatial local characteristics of the intensity. The inter-frame predictor is motion-adaptive using a pel-recursive method estimating the displacement vector. Weight coefficients are continuously adapted in order to favor the prediction mode which performs better between intra-frame and motion compensation mode. It is a backwards adaptation which does not necessitate the transmission of an overhead information. Neither the weight coefficients nor displacement vectors are transmitted. Apart from the quantized prediction error, it may be necessary to transmit the detection of a discontinuity of the displacement vector. For the examined image sequence a significant improvement is obtained in comparison with only adaptive intra-frame or only motion compensation mode. We give and discuss the extension of a known adaptive quantizer for 2D signals. A crucial problem in predictive coding, particularly with adaptive techniques, is the sensitivity to transmission errors. A method ensuring the self-adjustment of the decoder in the presence of transmission errors, which do not affect the pixel synchronization, is proposed for the intra-frame mode. Neither overhead information nor error-correcting codes are needed.     

基于动态时间规整和主动外观模型的动态表情识别

针对静态表情特征缺乏时间信息,不能充分体现表情的细微变化,该文提出一种针对非特定人的动态表情识别方法：基于动态时间规整(Dynamic Time Warping, DTW)和主动外观模型(Active Appearance Model, AAM)的动态表情识别。首先采用基于局部梯度DT-CWT(Dual-Tree Complex Wavelet Transform)主方向模式(Dominant Direction Pattern, DDP)特征的DTW对表情序列进行规整。然后采用AAM定位出表情图像的66个特征点并进行跟踪,利用中性脸的特征点构建人脸几何模型,通过人脸几何模型的匹配克服不同人呈现表情的差异,并通过计算表情序列中相邻两帧图像对应特征点的位移获得表情的变化特征。最后采用最近邻分类器进行分类识别。在CK+库和实验室自建库HFUT-FE(HeFei University of Technology-Face Emotion)上的实验结果表明,所提算法具有较高的准确性。     

Efficient facial landmark localization using spatial–contextual AdaBoost algorithm

Facial landmark detectors can be categorized into global and local detectors. Global facial landmark detectors rely on global statistical relations between landmarks, but do not sufficiently utilize local appearance information, whereas local detectors mainly focus on local appearance attributes of landmarks. Although the AdaBoost algorithm has been successfully employed in object localization, it cannot take advantage of geometric facial feature distribution very well. We propose an AdaBoost algorithm called SC-AdaBoost, which efficiently combines the global knowledge of landmark distribution, the regional shape model, and the local landmark attributes based on a coarse-to-fine strategy. The global prior distribution of landmarks is estimated using a face image set with landmark annotations. First, the face region is detected as a rectangular bounding box using a Haar-like feature-based boosting method, and the global distribution of landmarks is used to determine the facial component regions. Facial landmark localization is roughly performed by regional shape modeling. Posteriors of individual weak classifiers are determined by Gabor wavelet analysis at landmark candidate positions constrained by the regional shape model. SC-AdaBoost is established by empirical risk minimization, which decides the weights for the weak classifiers, and is used for the precise localization. The strength of the proposed approach is shown by extensive experiments using standard face datasets.     

模型基编码的运动参数估计及误差准则

在人脸序列的图像编码中 ,模型基编码方法可以获得高的主观图像质量和低的码率 ,而受到广泛重视。但是 ,其运动参数的可靠估计还是一个难点 ,而且也没有一个较好的适合视觉特性的误差准则。本文提出了基于特征点的运动参数估计算法 ,并根据边沿 ,亮度和端点特性来自动提取特征点及自适应调整点的数目。提出用重建的图像的质量来估价运动参数误差 ,并给出了误差面积和轮廓转折率误差二个函数。这二个函数较好地反映了运动参数误差引入的图像几何失真。     

基于深度学习与特征融合的人脸识别算法

卷积神经网络在人脸识别研究上有较好的效果,但是其提取的人脸特征忽略了人脸的局部结构特征。针对此问题,文中提出一种基于深度学习与特征融合的人脸识别方法。该算法将局部二值模式信息与原图信息相结合作为SDFVGG网络的输入,使得提取的人脸特征更加丰富且更具表征能力。其中,SDFVGG网络是将VGG网络进行深浅特征相融合后的网络。在CAS-PEAL-R1人脸数据库上的实验表明,将网络深浅特征相融合与在卷积神经网络中加入LBP图像信息与原图信息相融合的特征信息对于提高人脸识别准确率非常有效,可得到优于传统算法和一般卷积神经网络的最高98.58%人脸识别率。