基于局部谱能量自相似矩阵的纹理描述

对于纹理检测和分类中的纹理描述问题,提出一种新的基于Gabor滤波器组局部谱能量的自相似矩阵来描述纹理的方法。首先采用多尺度、方向的极坐标对数Gabor滤波器组对纹理模板进行滤波,获得频域上局部频段和方向上的纹理信息;然后计算频域上各尺度、方向上局部谱能量的自相似度量,将这些度量值以自相似矩阵的形式进行存储,并作为纹理特征的描述子;最后将这种描述方法应用到纹理检测和分类中。由于该描述子主要体现的是纹理模板在不同频段和方向局部谱能量的自相似程度,所以它对滤波器参数的依赖度较低。实验中利用纹理特征描述子可以实现比较准确的纹理检测,多类纹理合成图像分类实验的准确率达到了91%以上。实验结果说明,纹理局部谱能量的自相似矩阵是一种十分有效的纹理描述方法,其检测和分类的结果对后期的纹理分割、纹理识别等研究领域具有广泛的应用前景。     

Deep cross-view autoencoder network for multi-view learning

In many real-world applications, an increasing number of objects can be collected at varying viewpoints or by different sensors, which brings in the urgent demand for recognizing objects from distinct heterogeneous views. Although significant progress has been achieved recently, heterogeneous recognition (cross-view recognition) in multi-view learning is still challenging due to the complex correlations among views. Multi-view subspace learning is an effective solution, which attempts to obtain a common representation from downstream computations. Most previous methods are based on the idea of maximal correlation after feature extraction to establish the relationship among different views in a two-step manner, thus leading to performance deterioration. To overcome this drawback, in this paper, we propose a deep cross-view autoencoder network (DCVAE) that extracts the features of different views and establishes the correlation between views in one step to simultaneously handle view-specific, view-correlation, and consistency in a joint manner. Specifically, DCVAE contains self-reconstruction, newly designed cross-view reconstruction, and consistency constraint modules. Self-reconstruction ensures the view-specific, cross-view reconstruction transfers the information from one view to another view, and consistency constraint makes the representation of different views more consistent. The proposed model suffices to discover the complex correlation embedded in multi-view data and to integrate heterogeneous views into a latent common representation subspace. Furthermore, the 2D embeddings of the learned common representation subspace demonstrate the consistency constraint is valid and cross-view classification experiments verify the superior performance of DCVAE in the two-view scenario.

     

Cross-view action recognition by cross-domain learning

This paper proposes a novel cross-view human action recognition method by discovering and sharing common knowledge among different video sets captured in multiple viewpoints. We treat a specific view as target domain and the others as source domains and consequently formulate the cross-view action recognition into the cross-domain learning framework. First, the classic bag-of-visual word framework is implemented for visual feature extraction in individual viewpoints. Then, we add two transformation matrices in order to transform original action feature from different views into one common feature space, and also combine the original feature and the transformation feature to proposed the new feature mapping function for target and auxiliary domains respectively. Finally, we proposed a new method to learn the two transformation matrices in model training step based on the standard SVM solver and generate the final classifier for each human action. Extensive experiments are implemented on IXMAS, and TJU. The experimental results demonstrate that the proposed method can consistently outperform the state-of-the-arts.     

基于多视图半监督学习的人体行为识别

由于人的行为在本质上的复杂性,单一行为特征视图缺乏全面分析人类行为的能力.文中提出基于多视图半监督学习的人体行为识别方法.首先,提出3种不同模态视图数据,用于表征人体动作,即基于RGB模态数据的傅立叶描述子特征视图、基于深度模态数据的时空兴趣点特征视图和基于关节模态数据的关节点投影分布特征视图.然后,使用多视图半监督学习框架建模,充分利用不同视图提供的互补信息,确保基于少量标记和大量未标记数据半监督学习取得更好的分类精度.最后,利用分类器级融合技术并结合3种视图的预测能力,同时有效解决未标记样本置信度评估问题.在公开的人体行为识别数据集上实验表明,采用多个动作特征视图融合的特征表示方法的判别力优于单个动作特征视图,取得有效的人体行为识别性能.     

MMA: a multi-view and multi-modality benchmark dataset for human action recognition

Human action recognition is an active research topic in both computer vision and machine learning communities, which has broad applications including surveillance, biometrics and human computer interaction. In the past decades, although some famous action datasets have been released, there still exist limitations, including the limited action categories and samples, camera views and variety of scenarios. Moreover, most of them are designed for a subset of the learning problems, such as single-view learning problem, cross-view learning problem and multi-task learning problem. In this paper, we introduce a multi-view, multi-modality benchmark dataset for human action recognition (abbreviated to MMA). MMA consists of 7080 action samples from 25 action categories, including 15 single-subject actions and 10 double-subject interactive actions in three views of two different scenarios. Further, we systematically benchmark the state-of-the-art approaches on MMA with respective to all three learning problems by different temporal-spatial feature representations. Experimental results demonstrate that MMA is challenging on all three learning problems due to significant intra-class variations, occlusion issues, views and scene variations, and multiple similar action categories. Meanwhile, we provide the baseline for the evaluation of existing state-of-the-art algorithms.     

基于生成对抗网络的跨视角步态特征提取

步态是一种能够在远距离、非侵犯的条件下识别身份的生物特征,但在实际场景中,步态很容易受到拍摄视角、行走环境、物体遮挡、着装等因素的影响.在跨视角识别问题上,现有方法只注重将多种视角的步态模板转化到固定视角下,且视角跨度的增大加深了错误的累积.为了提取有效的步态特征用于跨视角步态识别,本文提出了一种基于生成对抗网络的跨视角步态特征提取方法,该方法只需训练一个模型即可将步态模板转换到任意视角下的正常行走状态,并最大化地保留原本的身份特征信息,从而提高步态识别的准确率.在CASIA-B和OUMVLP数据集上的实验结果表明,该方法在解决跨视角步态识别问题上具有一定的鲁棒性和可行性.     

共享转换矩阵的胶囊网络及其融合视角特征的跨视角步态识别

步态识别具有非接触性、非侵犯性、易感知等优势,然而,在跨视角的步态识别中,行人的轮廓会随人的视角的变化而不同,从而影响步态识别的性能。为此,提出了共享转换矩阵的胶囊网络及其改进的动态路由算法,从而减少了网络训练参数。在此基础上,通过融合视角特征,利用Triplet损失与Margin损失提出了融合视角特征的跨视角步态识别模型。在CASIA-B数据集上的实验结果表明,使用共享转换矩阵的胶囊网络提取步态特征是有效的,在正常行走、携带背包、穿戴外套条件下,所提融合视角特征的模型在识别准确率上比基于卷积神经网络的跨视角步态识别方法提高了4.13%,且对跨较大视角的步态识别具有更好的性能。     

基于双通道循环一致性GAN的跨视角步态识别研究

步态识别系统在生物识别领域显示出巨大的潜力,然而步态识别的准确性很容易受到视角的影响。为解决这一问题,提出的方法基于循环生成对抗网络(cycle generative adversarial network, Cycle-GAN)的网络结构,结合新的特征提取模块以及多重损失函数,提出了一种用于跨视角步态识别的网络模型双通道循环一致性生成对抗网络(two-channel cycle consistency generative adversarial network, TCC-GAN)。该网络首先将步态能量图像从任意视角转换为目标视角图像,然后进行比对从而实现身份识别。TCC-GAN分别引入了重建损失、视角分类和身份保持损失来指导生成器生成具有目标视角的步态图像并同时保留身份信息。为了避免可能存在的模式崩塌问题,并保证各个输入和输出以有意义的方式进行映射,模型中还利用了循环一致性损失。数据集CASIA-B和OU-MVLP上的实验结果表明:所提TCC-GAN模型的跨视角识别率高于目前大多数其他基于GAN的跨视角步态识别模型。