基于HOURGLASS网络语义关键点提取的光学图像空间目标姿态估计方法

空间目标姿态估计是有效实现各类航天任务的重要前提,基于空间光学观测图像的目标姿态估计关键一环在于快速准确地建立起观测图像与空间目标之间的“二维特征点-三维实体结构”映射关系。传统的方法往往将这一任务分解为特征提取和特征关联两个步骤序贯进行,然而在空间目标光学观测场景中,高动态的光照变化和目标的相对高速运动特点会显著降低图像特征提取的可靠性,影响后续特征关联匹配的正确率并最终降低对空间目标的姿态估计精度。针对这一问题,本文提出了一种基于语义关键点提取的光学图像空间目标姿态估计方法,利用Hourglass网络端到端地提取包含语义信息的关键点,直接实现了光学图像中二维特征点与目标三维实体结构的关联映射,并在此基础上利用EPnP算法求解待估计的目标姿态值。实验结果表明,本文所提的方法能较好地兼顾算法精度与效率,其在仿真数据集上的姿态估计最小误差为0.83°,且在数据降质的情况下平均误差依然优于传统方法。      

基于瞬时步态能量图的远距离身份识别

提出了一种基于瞬时步态能量图的远距离身份识别算法.首先根据摆动距离计算出步态周期,并指定步态周期中的关键时刻.步态序列中一个关键时刻的所有侧面轮廓图的平均值构成一个平均瞬时图.一个关键时刻的瞬时步态能量图的计算利用了当前关键时刻以及其他关键时刻的平均瞬时图.提高了每个关键时刻侧面轮廓图像的质量,并比单纯使用步态能量图的方式增加了步态的运动信息.随后计算出所有关键时刻侧面轮廓图相对于瞬时步态能量图的偏移的累积图像,与步态能量图共同作为描述一个对象的特征向量.最后,使用最近邻算法进行步态特征分类.在USF步态数据库上对该算法进行实验,并与基线算法以及另外两个新的步态识别算法进行比较,结果显示该算法达到了更高的总体识别率.     

Pose Depth Volume extraction from RGB-D streams for frontal gait recognition

We explore the applicability of Kinect RGB-D streams in recognizing gait patterns of individuals. Gait energy volume (GEV) is a recently proposed feature that performs gait recognition in frontal view using only depth image frames from Kinect. Since depth frames from Kinect are inherently noisy, corresponding silhouette shapes are inaccurate, often merging with the background. We register the depth and RGB frames from Kinect to obtain smooth silhouette shape along with depth information. A partial volume reconstruction of the frontal surface of each silhouette is done and a novel feature termed as Pose Depth Volume (PDV) is derived from this volumetric model. Recognition performance of the proposed approach has been tested on a data set captured using Microsoft Kinect in an indoor environment. Experimental results clearly demonstrate the effectiveness of the approach in comparison with other existing methods.     

基于Gabor相位谱和流型学习的步态识别方法

 提出了一种有效的基于步态能量图像的身份识别方法.首先生成合成步态能量图像(GEI)丰富训练集样本数量.然后利用在以前文献中被忽略的具有良好识别性能的Gabor相位信息作为身份特征,并采用流型学习算法保局影射(LPP)将此高维数据在低维空间表示.通过使用简单的分类策略在USF步态数据库上进行对比实验,结果表明本方法的正确识别率优于现有其他的自动步态识别算法.     

Gait identification with partial occlusion using six modules and consideration of occluded module exclusion

In gait identification, partial occlusion sometimes occurs and leads to missed identification. This paper proposes a gait identification method for partial occlusion case by using six modules and consideration of occluded module exclusion. In this method, a Gait Energy Image (GEI) is separated into four individual modules, and three of neighboring modules are coupled into other two coupling modules. When partial occlusion of a module occurs, the occluded module is detected and excluded from consideration for gait identification. In addition, the combined TDPCA and TDLDA are employed to extract gait features comparing with trained features in the database, and the candidate with the highest score in matching with the database is selected as identified person. To evaluate performance of proposed method, experiments carried out with CASIA dataset with 123 classes and our own EEPIT dataset with 135 classes indicate effectiveness of module separation and significance of exclusion of occluded modules.     

Human gait recognition based on Haralick features

This paper proposes a supervised feature extraction approach that is capable of selecting distinctive features for the recognition of human gait under clothing and carrying conditions, thus improving the recognition performances. The principle of the suggested approach is based on the Haralick features extracted from gait energy image (GEI). These features are extracted locally by dividing vertically or horizontally the GEI locally into two or three equal regions of interest, respectively. RELIEF feature selection algorithm is then employed on the extracted features in order to select only the most relevant features with a minimum redundancy. The proposed method is evaluated on CASIA gait database (Dataset B) under variations of clothing and carrying conditions for different viewing angles, and the experimental results using k-NN classifier have yielded attractive results of up to 80% in terms of highest identification rate at rank-1 when compared to existing and similar state-of-the-art methods.     

Gabor phase based gait recognition

An effective gait recognition approach based on the gait energy image (GEI) representation is proposed. Synthetic GEIs are first created to address the problem of lacking training data. Then the Gabor phase spectrum of the GEI which was ignored in previous works is utilised as an input feature, and it is subsequently projected into a low dimensional space by linear discriminant analysis to perform classification. Experimental results show that the proposed approach outperforms other algorithms in terms of recognition accuracy.     

姿态估计和跟踪结合的运动视频关键帧提取

提出姿态估计和特定部位跟踪相结合的动作视频关键帧提取算法.首先利用非确定人体部位的时间连续性保持提高基于柔性部件铰接人体模型的单帧图像人体姿态估计准确率,通过实施数据降维得到局部拓扑结构表达能力强的判别性运动特征向量,采用极值判定原理确定候选关键帧集合.然后利用ISODATA动态聚类算法,通过初始聚类中心优化、基于语义的关键帧集合增强等策略确定关键帧.实验表明文中算法具有较高的关键帧提取准确率和召回率,支持基于语义的关键帧提取.提取的视频关键帧可以用于运动视频压缩和批注审阅.     

复杂场景点云数据的6D位姿估计深度学习网络

针对工业上常见的弱纹理、散乱摆放复杂场景下点云目标机器人抓取问题,该文提出一种6D位姿估计深度学习网络。首先,模拟复杂场景下点云目标多姿态随机摆放的物理环境,生成带真实标签的数据集;进而,设计了6D位姿估计深度学习网络模型,提出多尺度点云分割网络(MPCS-Net),直接在完整几何点云上进行点云实例分割,解决了对RGB信息和点云分割预处理的依赖问题。然后,提出多层特征姿态估计网(MFPE-Net),有效地解决了对称物体的位姿估计问题。最后,实验结果和分析证实了,相比于传统的点云配准方法和现有的切分点云的深度学习位姿估计方法,所提方法取得了更高的准确率和更稳定性能,并且在估计对称物体位姿时有较强的鲁棒性。     

Synchronized Submanifold Embedding for Person-Independent Pose Estimation and Beyond

Precise 3-D head pose estimation plays a significant role in developing human-computer interfaces and practical face recognition systems. This task is challenging due to the particular appearance variations caused by pose changes for a certain subject. In this paper, the pose data space is considered as a union of submanifolds which characterize different subjects, instead of a single continuous manifold as conventionally regarded. A novel manifold embedding algorithm dually supervised by both identity and pose information, called snchronized submanifold embedding (SSE), is proposed for person-independent precise 3-D pose estimation, which means that the testing subject may not appear in the model training stage. First, the submanifold of a certain subject is approximated as a set of simplexes constructed using neighboring samples. Then, these simplexized submanifolds from different subjects are embedded by synchronizing the locally propagated poses within the simplexes and at the same time maximizing the intrasubmanifold variances. Finally, the pose of a new datum is estimated as the propagated pose of the nearest point within the simplex constructed by its nearest neighbors in the dimensionality reduced feature space. The experiments on the 3-D pose estimation database, CHIL data for CLEAR07 evaluation, and the extended application for age estimation on FG-NET aging database, demonstrate the superiority of SSE over conventional regression algorithms as well as unsupervised manifold learning algorithms.      

基于加强步态能量图在非规范视角的步态识别研究

提出了一种基于加强步态能量图的非规范视角步态识别方法,解决了非规范视角下步态识别难题。视角转换方法将视角统一,采用背景减除法提取人体轮廓,引入步态周期检测方法确定步态周期,根据人体骨架参数模型得到加强步态能量图（E-GEI）,最后运用2DPCA方法提取特征向量,并采用最近邻域法分类。实验结果表明：E-GEI在各个视角下比普通的GEI在识别效果要更好。     

An Efficient Image Forgery Detection Using Biorthogonal Wavelet Transform and Improved Relevance Vector Machine

Nowadays, the development of refined image processing and software editing tools has finish the exploitation of digital images easily and invisible the image to the normal eyes and this process known as image fakery. Image security is one of the key issues in any field that makes use of digital images. Copy-move forgery (CMF) is the most effective and simple scheme to create forged digital images. In general, the methodologies based on Scale Invariant Feature Transform (SIFT) are widely used to detect CMF. Unfortunately, the detection performance of all SIFT based CMF detection approaches are extremely dependent on the selection of feature vectors. The values of these parameters are often determined through experience or some experiments on a number of forgery images. However, these experience parameter values are not applicable to every image thereby offers a limited usefulness. This paper deals the CMF problem using improved Relevance Vector Machine technique. The key idea of the IVRM is to apply Biorthogonal Wavelet Transform based scheme on image for feature extraction. The feature vectors are then stored lexicographically and similarity of vectors is decided using Minkowski distance and threshold value. The simulation results of proposed technique show a significant improvement in accuracy, sensitivity, and specificity rates over others existing schemes.     

Reducing covariate factors of gait recognition using feature selection and dictionary-based sparse coding

Human gait recognition is a behavioral biometrics method that aims to determine the identity of individuals through the manner and style of their distinctive walk. It is still a very challenging problem because natural human gait is affected by many covariate factors such as changes in the clothing, variations in viewing angle, and changes in carrying condition. This paper evaluates the most important features of gait under the carrying and clothing conditions. We find that the intra-class variations of the features that remain static during the gait cycle affect the recognition accuracy adversely. Thus, we introduce an effective and robust feature selection method based on the gait energy image. The new gait representation is less sensitive to these covariate factors. We also propose an augmentation technique to overcome some of the problems associated with the intra-class gait fluctuations, as well as if the amount of the training data is relatively small. Finally, we use dictionary learning with sparse coding and linear discriminant analysis to seek the best discriminative data representation before feeding it to the Nearest Centroid classifier. When our method is applied on the large CASIA-B gait data set, we are able to outperform existing gait methods by achieving the highest average result.     

Driver activity recognition by learning spatiotemporal features of pose and human object interaction

Detecting hazardous activity during driving can be useful in curbing roadside accidents. Existing techniques utilizing image based features for encoding such activity can sometimes misclassify crucial scenarios. One particular work by Zhao et al. (2013 [1], 2013 [2], 2011 [3]) suggests an image based feature set that encodes the driver’s pose, which is categorized into one of four activities. We bring more clarity in understanding the activity by proposing a richer, video based feature set that adeptly exploits spatiotemporal information of the driver. Our feature set encodes the driver’s pose, crucial variations in pose and interactions with objects within the vehicle. The feature set is tested on our newly created dataset since the ones used in literature are not publicly available. Our proposed feature set captures a larger number of activities and using standard classifiers and benchmarks it has shown significant improvements over the existing ones.     

Model-based temporal object verification using video

An approach to model-based dynamic object verification and identification using video is proposed. From image sequences containing the moving object, we compute its motion trajectory. Then we estimate its three-dimensional (3-D) pose at each time step. Pose estimation is formulated as a search problem, with the search space constrained by the motion trajectory information of the moving object and assumptions about the scene structure. A generalized Hausdorff (1962) metric, which is more robust to noise and allows a confidence interpretation, is suggested for the matching procedure used for pose estimation as well as the identification and verification problem. The pose evolution curves are used to assist in the acceptance or rejection of an object hypothesis. The models are acquired from real image sequences of the objects. Edge maps are extracted and used for matching. Results are presented for both infrared and optical sequences containing moving objects involved in complex motions     

A Novel Image Mosaicking Technique for Enlarging the Field of View of Images Transmitted over Wireless Image Sensor Networks

Existing image mosaicking algorithms generate a complete scene that incorporates a number of images captured by several cameras. The traditional image mosaicking approaches cannot be applied directly to the emerging Wireless Image Sensor Networks (WISNs), since the low performance of image transmission over wireless sensor networks causes a noticeable delay before an entire image is received by a control center node. In this work, we propose a Progressive Image Mosaicking Algorithm (PIMA) based on the multi-scan feature of Progressive JPEG (P-JPEG). The originality of PIMA is based essentially on how it successfully performs mosaicking by using incremental image quality, as opposed to traditional methods that require complete data from all images. PIMA builds mosaics of images that are decoded from P-JPEG scans at three levels of quality, and delivers an approximate view of the scene in a short time while the reception of further image data is still in progress. Thereafter, it updates the image registration on two other refined levels to gradually enhance the display quality. We also propose the concept of Richer Information and Likeliest (RIL) block pair, which is a variation of the Sum of Absolute Difference (SAD). RIL can improve significantly the accuracy of image registration. We have conducted an extensive set of experiments and evaluated our proposed schemes against selected existing approaches. Our performance results indicate that PIMA decreases the delay before the first display of the scene, while preserving equivalent performance and image quality when compared to existing patch-based image mosaicking algorithms.     

Multi-manifold-based skin classifier on feature space Voronoï regions for skin segmentation

Skin segmentation is a crucial and a challenging step in many face and gesture recognition techniques and it has various applications in human computer interaction, objectionable content filtering, image retrieval and many more. In this article, we propose a novel skin segmentation method, which uses multi-manifold-based skin classification of feature space skin candidate Voronoï regions to achieve accurate skin segmentation. The state-of-the-art skin segmentation techniques reported in this article focus on discrimination between textural feature vectors belonging to skin and non-skin classes. In contrast, the proposed method focuses on discrimination between textural feature vectors belonging to skin and skin-like (non-skin) classes, which lead to higher skin classification accuracy. Furthermore, we introduce a novel image segmentation technique based on spatial and feature space Dirichlet tessellation (also called a Voronoï diagram) to achieve feature space segmentation of skin candidate regions of an image. These feature space segments will then be classified using a multi-manifold-based skin classifier. The proposed skin segmentation method was evaluated on two benchmark skin segmentation data sets and its results were compared with four other state-of-the-art methods proposed for skin segmentation. The experimental results reported in this article confirm that the proposed method outperforms the existing skin segmentation approaches in terms of false alarm rates in the skin segmentation process. Also, the proposed method results in the lowest minimal detection error compared to the existing methods reported in this article.     

基于线特征差分投影的空间目标姿态估计方法

空间目标姿态估计是有效实现基于ISAR图像空间目标识别的重要前提。本文针对利用线特征二维投影进行姿态估计时,线特征投影的检测误差会严重影响姿态估计精度这一问题,提出一种基于线特征差分投影的空间目标姿态估计方法。该法利用DP算法检测线特征在ISAR图像中的投影,通过建立线特征在实测ISAR图像和姿态估计值下的仿真图像中的二维差分投影,将线特征投影检测的绝对误差转化成相对误差,有效减小了线特征投影的检测误差对姿态估计的影响;同时,利用差分投影求取姿态估计的修正量,形成姿态估计的优化迭代过程,不断提高姿态估计精度。仿真实验验证了方法的可靠性与有效性。      

基于多核稀疏编码的三维人体姿态估计

为了准确有效的重构多视角图像中的三维人体姿态,该文提出一种基于多核稀疏编码的人体姿态估计算法.首先,针对连续帧姿态估计的歧义问题,该文设计了一种用于表达多视角图像的HA-SIFT描述子,其中,人体局部拓扑、肢体相对位置及外观信息被同时编码;然后,在多核学习框架下建立同时考虑特征空间内在流形结构与姿态空间几何信息的目标函数,并在希尔伯特空间优化目标函数以更新稀疏编码、过完备字典与多核权值;最后,利用姿态字典原子的线性组合来估计对应未知输入的三维人体姿态.实验结果表明,与核稀疏编码、Laplace稀疏编码及Bayesian稀疏编码相比,文本方法具有更高的估计精度.     

Face Recognition Using 3-D Models: Pose and Illumination

Unconstrained illumination and pose variation lead to significant variation in the photographs of faces and constitute a major hurdle preventing the widespread use of face recognition systems. The challenge is to generalize from a limited number of images of an individual to a broad range of conditions. Recently, advances in modeling the effects of illumination and pose have been accomplished using three-dimensional (3-D) shape information coupled with reflectance models. Notable developments in understanding the effects of illumination include the nonexistence of illumination invariants, a characterization of the set of images of objects in fixed pose under variable illumination (the illumination cone), and the introduction of spherical harmonics and low-dimensional linear subspaces for modeling illumination. To generalize to novel conditions, either multiple images must be available to reconstruct 3-D shape or, if only a single image is accessible, prior information about the 3-D shape and appearance of faces in general must be used. The 3-D Morphable Model was introduced as a generative model to predict the appearances of an individual while using a statistical prior on shape and texture allowing its parameters to be estimated from single image. Based on these new understandings, face recognition algorithms have been developed to address the joint challenges of pose and lighting. In this paper, we review these developments and provide a brief survey of the resulting face recognition algorithms and their performance