基于DBN-HMM的人体动作识别

动作识别使得机器能够对人体动作的意图进行判别理解，进而实现高效的人机交互。提出一种肢体角度模型，实现在三维空间中对人体动作进行表示，该模型具有一定的不变性，计算复杂度低。针对传统的基于混合高斯的隐马尔可夫模型（GMM-HMM）的动作识别，提出深度置信网络模型（DBN）和隐马尔可夫模型相结合的动作识别模型，构建了一种非线性的基于条件限制玻尔兹曼机（CRBM）的DBN深度学习模型，深层次结构使其建模能力更强，且能够结合历史信息建模，更适用于动作识别。实验表明该算法具有较高的识别结果。     

ADR-SPLDA: Activity discovery and recognition by combining sequential patterns and latent Dirichlet allocation

This paper presents ADR-SPLDA, an unsupervised model for human activity discovery and recognition in pervasive environments. The activities are encoded in sequences recorded by non-intrusive sensors placed at various locations in the environment. Our model studies the relationship between the activities and the sequential patterns extracted from the sequences. Activity discovery is formulated as an optimization problem in which sequences are modeled as probability distributions over activities, and activities are, in turn, modeled as probability distributions over sequential patterns. The optimization problem is solved by maximization of the likelihood of data. We present experimental results on real datasets gathered in smart homes where people perform various activities of daily living. The results obtained demonstrate the suitability of our model for activity discovery and characterization. Also, we empirically demonstrate the effectiveness of our model for activity recognition by comparing it with two of the widely used models reported in the literature, the Hidden Markov model and the Conditional Random Field model.     

基于条件随机域的上下文人类动作识别

提出一种新的基于条件随机域和隐马尔可夫模型（HMM）的人类动作识别方法——HMCRF。目前已有的动作识别方法均使用隐马尔可夫模型及其变型,这些模型一个最突出的不足就是要求观察值相互独立。条件模型很容易表示上下文相关性,且可使用动态规划做到有效且精确的推论,它的参数可以通过凸函数优化训练得到。把条件图形模型应用于动作识别之上,并通过大量的实验表明,所提出的方法在识别正确率方面明显优于一般线性结构的CRF和HMM。     

一种基于BP—HMM的字符识别方法

传统隐马尔柯夫模型广泛地应用在字符识别中,并具有较强的识别能力,但不能兼顾每个模型对其对应目标有很强的识别能力和模型之间差异性的最大化。该文提出的BP—隐马尔柯夫模型通过训练样本的不断训练,调整自身参数,解决了传统隐马尔柯夫模型不能解决的问题。计算机仿真结果表明:BP—隐马尔柯夫模型较传统的隐马尔柯夫模型有更强的抗干扰能力和更高的字符识别率。     

A comparative study of two state-of-the-art sequence processing techniques for hand gesture recognition

In this paper, we address the problem of the recognition of isolated, complex, dynamic hand gestures. The goal of this paper is to provide an empirical comparison of two state-of-the-art techniques for temporal event modeling combined with specific features on two different databases. The models proposed are the Hidden Markov Model (HMM) and Input/Output Hidden Markov Model (IOHMM), implemented within the framework of an open source machine learning library (www.torch.ch). There are very few hand gesture databases available to the research community; consequently, most of the algorithms and features proposed for hand gesture recognition are not evaluated on common data. We thus propose to use two publicly available databases for our comparison of hand gesture recognition techniques. The first database contains both one- and two-handed gestures, and the second only two-handed gestures.     

一种基于HMM的场景识别方法

隐马尔科夫模型作为一种统计分析模型,能够通过观测向量序列计算其隐含状态的概率分布密度。提出一种智能空间中基于HMM的场景识别方法,该方法指定系统相关情境信息,确定隐含场景集和观察情境集,采用部分相关情境信息而非全部情境信息作为场景特征参与场景识别,利用HMM对隐含场景间的关系进行建模,设计了基于HMM的场景识别算法。实验结果表明,采用基于HMM的场景识别方法能够获得较高的识别效率。     

汉语连续语音中HMM模型状态数优化方法研究

为了优化汉语连续语音中HMM模型系统以提高识别性能,提出了分别为每个声母和韵母半音节声学模型选择最优的状态数的方法。通过综合考虑每个声母和韵母半音节声学模型在不同状态数下的段长均值、方差以及各自识别率这三者信息,作为进行最优模型状态数的选择准则。优化后的声学模型系统由状态数各不相同的声母半音节声学模型组成,同未优化前状态数统一的模型系统相比,音节识别性能提高了5.07个百分点。研究表明,每个声母和韵母半音节志学模型应根据情况选择不同的状态数,优化后的模型系统识别性能得到了提高。     

隐马尔可夫模型及其最新应用与发展①

隐马尔可夫模型是序列数据处理和统计学习的一种重要概率模型,已被成功应用于许多工程任务中。首先介绍了隐马尔可夫模型的基本原理,接着综述了其在人的行为分析、网络安全和信息抽取中的最新应用。最后对最近提出来的无限状态隐马尔可夫模型的原理及最新发展进行了总结。     

隐马尔可夫模型及其最新应用与发展①

隐马尔可夫模型是序列数据处理和统计学习的一种重要概率模型,已被成功应用于许多工程任务中。首先介绍了隐马尔可夫模型的基本原理,接着综述了其在人的行为分析、网络安全和信息抽取中的最新应用。最后对最近提出来的无限状态隐马尔可夫模型的原理及最新发展进行了总结。     

A smart segmentation technique towards improved infrequent non-speech gestural activity recognition model

Infrequent Non-Speech Gestural Activities (IGAs) such as coughing, deglutition and yawning help identify fine-grained physiological symptoms and chronic psychological conditions which are not directly observable from traditional daily activities. We propose a new wearable smart earring which is capable of differentiating IGAs in daily environment with single integrated accelerometer sensor signal processing. Our prior framework, GeSmart, shows significant improvement in IGAs recognition based on smart earring which necessitates users to replace the earring batteries frequently due to its energy hungry requirement (high sampling frequency) towards fine-grained IGAs recognition. In this improved work, we propose a new segmentation technique along with GeSmart which takes the advantages of change-point detection algorithm to segment sensor data streams, feature extraction and classification thus any machine learning technique can perform significantly well in low sampling rate. We also implement a baseline traditional graphical model based gesture recognition techniques and compare their performances with our model in terms of accuracy, energy consumption and degradation of sampling rate scenarios. Experimental results based on real data traces demonstrate that our approach improves the performances significantly compared to previously proposed solutions. We also apply our segmentation technique on two benchmark datasets to prove the superiority of our technique in low sampling rate scenario.     

基于三维加速度的连续手势识别

由于基于图像处理的手势识别方法对环境背景要求较高且存在不稳定性问题,文章使用三维加速度传感器的连续数据进行手势识别.三维加速度传感器内置于大部分智能手机中,具有应用方便的特点.实验通过传感器获取加速度信号,经过低通滤波、去重力和特征提取的信号预处理过程后,结合隐马尔可夫模型和混合高斯模型的理论方法,实现手机手势的连续识别,并驱动应用层预先定义的交互命令.     

Gaussian Process Dynamical Models for hand gesture interpretation in Sign Language

Classifying human hand gestures in the context of a Sign Language has been historically dominated by Artificial Neural Networks and Hidden Markov Model with varying degrees of success. The main objective of this paper is to introduce Gaussian Process Dynamical Model as an alternative machine learning method for hand gesture interpretation in Sign Language. In support of this proposition, the paper presents the experimental results for Gaussian Process Dynamical Model against a database of 66 hand gestures from the Malaysian Sign Language. Furthermore, the Gaussian Process Dynamical Model is tested against established Hidden Markov Model for a comparative evaluation. A discussion on why Gaussian Process Dynamical Model is superior over existing methods in Sign Language interpretation task is then presented.     

Semi-Markov conditional random fields for accelerometer-based activity recognition

Activity recognition is becoming an important research area, and finding its way to many application domains ranging from daily life services to industrial zones. Sensing hardware and learning algorithms are two important components in activity recognition. For sensing devices, we prefer to use accelerometers due to low cost and low power requirement. For learning algorithms, we propose a novel implementation of the semi-Markov Conditional Random Fields (semi-CRF) introduced by Sarawagi and Cohen. Our implementation not only outperforms the original method in terms of computation complexity (at least 10 times faster in our experiments) but also is able to capture the interdependency among labels, which was not possible in the previously proposed model. Our results indicate that the proposed approach works well even for complicated activities like eating and driving a car. The average precision and recall are 88.47% and 86.68%, respectively, which are higher than results obtained by using other methods such as Hidden Markov Model (HMM) or Topic Model (TM).     

基于便携式传感器数据的GMM-HMM人体姿态识别算法

人体姿态识别在人机交互, 游戏以及医疗健康等领域有着深远意义, 基于便携式传感器进行多种人体姿态高精度的稳定识别是该领域的研究难点. 本文采集了8种姿态的高频传感器数据, 提取原始数据的窗口时域特征组成数据集. 根据人体姿态的传感器数据特点将人体姿态划分为4个阶段, 使用高斯混合模型(Gaussian Mixture Model, GMM)拟合人体姿态的观测序列, 结合隐马尔可夫模型(Hidden Markov Model, HMM), 利用GMM-HMM算法进行姿态识别. 本文对比了不同窗口值下的一阶隐马尔可夫模型(1 Order Hidden Markov Model, 1OHMM)和二阶隐马尔可夫模型(2 Order Hidden Markov Model, 2OHMM)的效果, 当窗口值为8时, 2OHMM的性能最优, 整体召回率达到了95.30%, 平均准确率达到了95.23%. 与其它研究相比, 本文算法能识别的姿态种类较多, 算法识别性能较佳且算法耗时较短.     

Beyond Tracking: Modelling Activity and Understanding Behaviour

In this work, we present a unified bottom-up and top-down automatic model selection based approach for modelling complex activities of multiple objects in cluttered scenes. An activity of multiple objects is represented based on discrete scene events and their behaviours are modelled by reasoning about the temporal and causal correlations among different events. This is significantly different from the majority of the existing techniques that are centred on object tracking followed by trajectory matching. In our approach, object-independent events are detected and classified by unsupervised clustering using Expectation-Maximisation (EM) and classified using automatic model selection based on Schwarz's Bayesian Information Criterion (BIC). Dynamic Probabilistic Networks (DPNs) are formulated for modelling the temporal and causal correlations among discrete events for robust and holistic scene-level behaviour interpretation. In particular, we developed a Dynamically Multi-Linked Hidden Markov Model (DML-HMM) based on the discovery of salient dynamic interlinks among multiple temporal processes corresponding to multiple event classes. A DML-HMM is built using BIC based factorisation resulting in its topology being intrinsically determined by the underlying causality and temporal order among events. Extensive experiments are conducted on modelling activities captured in different indoor and outdoor scenes. Our experimental results demonstrate that the performance of a DML-HMM on modelling group activities in a noisy and cluttered scene is superior compared to those of other comparable dynamic probabilistic networks including a Multi-Observation Hidden Markov Model (MOHMM), a Parallel Hidden Markov Model (PaHMM) and a Coupled Hidden Markov Model (CHMM). First online version published in February, 2006     

HMM machine learning and inference for Activities of Daily Living recognition

In a human-centric smart space, Activities of Daily Living (ADL) analysis can provide very useful information for elder care and long-term care services. ADL is defined as an assessment of a person’s functional status. Many recent researches concentrate on designing a good Context Aware Computing System to automate the actions necessarily triggered by ADL recognitions. Implementing a correct ADL recognition engine is a hard work, but will repay the system with lower inference errors and higher system dependability. A good ADL recognition engine is required to adjust its inference strategy based on the learning capability in order to avoid a high error rate, especially in real world inputs with a significant difference as compared to those in the training phase. In this paper, we proposed a powerful inference engine based on the Hidden Markov Model, called the Adaptive Learning Hidden Markov Model (ALHMM), which combines the Viterbi and Baum–Welch algorithms to enhance the accuracy and learning capability. The assessments of ALHMM are conducted on the Python platform and show the practical feasibility of Activity Recognition in residential homes. Such a technique can provide the key answer required for advancing the state-of-the-art in context-aware computing and applications in real life.     

Learning object,grasping and manipulation activities using hierarchical HMMs

This article presents a probabilistic algorithm for representing and learning complex manipulation activities performed by humans in everyday life. The work builds on the multi-level Hierarchical Hidden Markov Model (HHMM) framework which allows decomposition of longer-term complex manipulation activities into layers of abstraction whereby the building blocks can be represented by simpler action modules called action primitives. This way, human task knowledge can be synthesised in a compact, effective representation suitable, for instance, to be subsequently transferred to a robot for imitation. The main contribution is the use of a robust framework capable of dealing with the uncertainty or incomplete data inherent to these activities, and the ability to represent behaviours at multiple levels of abstraction for enhanced task generalisation. Activity data from 3D video sequencing of human manipulation of different objects handled in everyday life is used for evaluation. A comparison with a mixed generative-discriminative hybrid model HHMM/SVM (support vector machine) is also presented to add rigour in highlighting the benefit of the proposed approach against comparable state of the art techniques.     

高可信度加权的多分类器融合行为识别模型

为了提高基于智能移动设备的人体日常行为识别准确率,提出一种高可信度加权的多分类器融合行为识别模型（MCFM）。针对不同智能设备内置加速度传感器获取的三轴加速度信息,优选出与人体行为相关度高的特征集作为该模型的输入,将决策树、支持向量机以及反向传播（BP）神经网络三个基分类器通过高可信度加权投票算（HRWV）法训练出一个新的融合分类器。实验结果表明,所提出的分类器融合模型能有效提高行为识别的准确率,对静止、散步、跑步、上楼及下楼五种日常行为的平均识别准确率达到94.88%。     

Independent shape component-based human activity recognition via Hidden Markov Model

In proactive computing, human activity recognition from image sequences is an active research area. In this paper, a novel human activity recognition method is proposed, which utilizes Independent Component Analysis (ICA) for activity shape information extraction from image sequences and Hidden Markov Model (HMM) for recognition. Various human activities are represented by shape feature vectors from the sequence of activity shape images via ICA. Based on these features, each HMM is trained and activity recognition is achieved by the trained HMMs of different activities. Our recognition performance has been compared to the conventional method where Principal Component Analysis (PCA) is typically used to derive activity shape features. Our results show that superior recognition is achieved with the proposed method especially for activities (e.g., skipping) that cannot be easily recognized by the conventional method. Furthermore, by employing Linear Discriminant Analysis (LDA) on IC features, the recognition results further improved significantly in the recognition performance.