Human Activity Analysis Research Articles

Joint movement similarities for robust 3D action recognition using skeletal data

Human action analysis based on 3D imaging is an emerging topic. This paper presents an approach for the problem of action recognition using information from a number of action descriptors calculated from a skeleton fitted to the body of a tracked subject. In the proposed approach, a novel technique that automatically determines discriminative sequences of relative joint positions for each action class is employed. In addition, we use an extended formulation of the longest common subsequence algorithm as a similarity function, which allows the classifier to reliably find the best match for extracted features from noisy skeletal data. The proposed approach is evaluated using two existing datasets from the literature, one captured using a Microsoft Kinect camera and the other using a motion capture system. The experimental results show that the approach outperforms existing skeleton-based algorithms in terms of its classification accuracy and is more robust in the presence of noise when compared to the dynamic time warping algorithm for human action recognition.

Soil weathering-water environment-ecological risks in Hanjiang River Basin, China

The objective of this study is to assess water environment and ecological risks from heavy metals of soils in Hanjiang River Basin, China. We carried out investigations on the changing patterns of spatiotemporal distribution of soil chemical elements, and further identified the changing behaviors, geochemical process and driving forces of heavy metals in soil by integrated analysis of natural environmental factors and human activities. In order to achieve these goals, we used field investigation, laboratory testing and GIS spatial simulation. The following findings were achieved. Firstly, general soil weathering in the Hanjiang River Basin basically evolves from Ca and Na stages to K stage. Ca and Na decreased rapidly, and silicate weathering evolved gradually. Major ions are subject to carbonate dissolution. About 2/3 of the soils in the basin are alkaline (pH > 7) which provide good conditions for mineralization of organic nitrogen. Secondly, specific soil weathering processes differ in spatial patterns and element transport, driven by integrated interior forces (e.g., mineral materials) and external forces (e.g., temperature, precipitation, wind, and gravity). Original mineral materials are changed into soils with different physical properties (soil texture, mechanical composition) and chemical components (organic matter). Upstream original mineral materials have been transported, accumulated, and deposited downstream, driven by forces of gravity, water transport, glacial effects, and wind. Thirdly, key ecological factors of plant growth change greatly. Increase of soil humus mitigates pollution hazards and enhances soil self-purification. Land use and cover is one of the key contributors to determine geochemical patterns of soil organic matter. Intensity of mineral weathering decreases with the decrease of biomass. The comprehensive pollution index and ecological risk index increased from northwest to southeast, corresponding to increased erosive rainfall and rainstorms more than human activities.

Relevance of Epistemological Pluralism for Resource Management Policies

The common believe of a single successful form of knowing the reality is misleading. A brief analysis of many human activities reveals a pluralistic perspective of several coexisting conceptual schemes that underlie the form in which women and men know and constitute the world in which we live. Here, I will discuss the necessity of taking seriously this pluralistic perspective when developing conceptual frameworks of human activities. As an example of great interest and importance, I will consider the case of resource management that has to contemplate the conservation and the response of ecosystems to human activities that may lead to irreversible damage. Ecosystems’ activity occurs over a wide range of space and time scales and, hence, policies and proposals of solutions to ecological problems should avoid narrow and monochromatic approaches. Specifically, the Pemon perspective of fire is a good example to show that for more robust and flexible resource management we need the knowledge and experience of different cultural and disciplinary groups.

Quantitative Analysis of Human Daily Activities Based on a Wireless Sensor System

Quantitative Analysis of Human Daily Activities Based on a Wireless Sensor System

Propagative Hough Voting for Human Activity Detection and Recognition

Generalized Hough voting (HV) has shown promising results in both object and action detection. However, most existing HV methods will suffer when insufficient training data are provided. We propose propagative HV to address this limitation and apply it to human activity analysis. Instead of training a discriminative classifier for local feature voting, we match individual local features to propagate the label and spatiotemporal configuration information of local features via HV. To enable a fast local feature matching, we index the local features using random projection trees (RPTs). RPTs can reveal the low-dimension manifold structure to provide adaptive local feature matching. Moreover, as the RPT index can be built in either labeled or unlabeled dataset, it can be applied to different tasks, such as activity search (limited training) and recognition (sufficient training). The superior performances on benchmarked datasets validate that our propagative HV can outperform state-of-the-art techniques in various activity analysis tasks, such as activity search, recognition, and prediction.

Human Movement Modeling and Activity Perception Based on Fiber-Optic Sensing System

This paper presents a flexible fiber-optic sensor-based pressure sensing system for human activity analysis and situation perception in indoor environments. In this system, a binary sensing technology is applied to reduce the data workload, and a bipedal movement-based space encoding scheme is designed to capture people's geometric information. We also develop a nonrepetitive encoding scheme to eliminate the ambiguity caused by the two-foot structure of bipedal movements. Furthermore, we propose an invariant activity representation model based on trajectory segments and their statistical distributions. In addition, a mixture model is applied to represent scenarios. The number of subjects is finally determined by Bayesian information criterion. The Bayesian network and region of interests are employed to facilitate the perception of interactions and situations. The results are obtained using distribution divergence estimation, expectation-maximization, and Bayesian network inference methods. In the experiments, we simulated an office environment and tested walk, work, rest, and talk activities for both one and two person cases. The experiment results have demonstrated that the average individual activity recognition is higher than 90%, and the situation perception rate can achieve 80%.

움직임 벡터와 GPU를 이용한 인간 활동성 분석

In this paper, We proposed the approach of GPU and motion vector to analysis the Human activity in real-time surveillance system. The most important part, that is detect blob(human) in the foreground. We use to detect Adaptive Gaussian Mixture, Weighted subtraction image for salient motion and motion vector. And then, We use motion vector for human activity analysis. In this paper, the activities of human recognize and classified such as meta-classes like this {Active, Inactive}, {Position Moving, Fixed Moving}, {Walking, Running}. We created approximately 300 conditions for the simulation. As a result, We showed a high success rate about 86~98%. The results also showed that the high resolution experiment by the proposed GPU-based method was over 10 times faster than the cpu-based method.

STFC: Spatio-temporal feature chain for skeleton-based human action recognition

Human action recognition and analysis has been of interest to researchers of computer vision for many years. This paper presents a method to recognize human actions from sequences of 3D joint positions. The major contributions of this paper include: (1) An action decomposition method that uses motion velocities, the direction of motion, and the curvatures of trajectories to encode the temporal decomposition of action into a sequence of meaningful atomic actions (actionlets); and (2) the concept of the Spatio-Temporal Feature Chain (STFC) that is introduced to represent the characteristic parameters of temporal sequential patterns, which exhibit greater robustness to noise and temporal misalignment. The effectiveness of the proposed method is evaluated on three challenging 3D action datasets captured by commodity depth cameras. The experimental evaluations show that the proposed approach achieves promising results compared to other state of the art algorithms.

Review on Vision based Human Activity Analysis

Recognizing human actions are important in various real time applications. Review on human activity analysis is provided in three sections. The first section in this paper presents an overall classification to Human activity analysis from feature extraction to recognition systems. In the second section a survey is included which provides technical information to activity analysis. Finally a brief description of databases which came across in survey is also included. The overall purpose of this paper is to provide a basic understanding to human activity analysis and to analyze the major challenge in human activity analysis. General Terms Human Activity Recognition (HAR), Feature Extraction, Feature Representation, Classification, Recognition

A potência da ação. Uma crítica ao naturalismo da violência

A tendência de algumas visões naturalistas de deduzir o sentido da ação humana da sua natureza biológica conduz a naturalizar a violência como fenômeno político inexorável. A violência naturalizada leva a sociedade para a dinâmica da inexorável violência, cuja única solução é a legitimação de uma violência total. Propomos, neste ensaio, a desconstrução crítica dos determinismos naturalistas por meio da análise da ação humana e da especificidade de sua potência que se manifesta como potência criadora (Cornelius Castoriadis) e potência do não (Giorgio Agamben). A potência da ação humana transcende os determinismos da sua natureza biológica, embora por eles esteja condicionada.

Pre-dispositional constitution and plastic disposition: toward a more adequate descriptive framework for the notions of habits, learning and plasticity

Behavioral studies and neurobiological models of mental illnesses can be used to inform theories of mind and action. In this paper I use specific aspects of some paradigmatic cases in order to establish what I consider to be a useful distinction for the analysis of human action and, more specifically, the delimitation of habitual action. Patient SM is a well-known case of Urbach-Wiethe disease—one of only 300 cases reported in the reference literature— that was submitted to a decade of investigations (Adolphs et al., 1994, 2005). This syndrome, also known as lipoid proteinosis, produces dermatological lesions as well as calcifications in regions of the brain, often affecting the amygdaloid region (Siebert et al., 2003; Bahadir et al., 2006). While her basic perception, memory, and language skills are essentially normal, SM has nearly complete bilateral destruction of the amygdala, and her social behavior is indiscriminately trusting and friendly (Adolphs et al., 1994). Ten years of research showed an intriguing impairment in her ability to recognize fear in facial expressions, due to a lack of spontaneous fixation on the eyes when viewing faces (Adolphs et al., 2005). The research showed that in control patients, spontaneous fixation is directed principally to the eyes and the mouth, tracking the regions of the face that allow one to distinguish facial expressions. However, patient SM spontaneously focused on the nose, thereby missing necessary information for judging emotions. What is so interesting to point out is that when given explicit instructions (“look at the eyes of this person”) SM had no problem focusing on the eyes and recognizing emotions but, surprisingly, after a decade of treatment, SM was not able to learn the habit of looking at the eyes of the face spontaneously. Another case of interest for action theory can be found in studies of subjects in the autistic spectrum (Klin et al., 2003; Boria et al., 2009; Gallese, 2009; Kana et al., 2014), particularly studies of deficits in the functioning of the “mirror mechanism” (Antonietti, 2013). This deficit appears related to other deficits such as atypical visual processing and encoding of social stimuli, as well as imitative behavior and the ability to share attention (for review, see Gallese et al., 2013). Atypical brain development has been identified in cases of autism; specifically, the neural organization in areas involved in social cognition, facial expression, and facial recognition, as well as in areas associated with the mirror mechanism, appears to be related to the functional architecture that characterizes the atypical development of the autistic spectrum (Cauda et al., 2011; Gallese et al., 2013). Another paradigmatic case that is informative for theories of mind-based action is schizophrenia (Synofzik et al., 2010; Leube et al., 2012; Mausbach et al., 2013). Multiple investigations suggest that schizophrenia and other neuropsychiatric disorders are associated with deficits in mirror neurons (Enticott et al., 2008; Mehta et al., 2013) and with interneuron dysfunction (Marin, 2012). For now, let us focus on the second dysfunction. Interneurons regulate the activity of pyramidal cells, largely through inhibitory mechanisms, and one of the functions of pyramidal cells is to maintain cerebral patterns associated with perception and memory. Migration of the interneurons during the development of the nervous system is fundamental to this function, as it determines the final positioning of the neurons, thereby establishing the basis for correct wiring of neural circuitry (Marin, 2013). It has been demonstrated that schizophrenics possess mutations in certain genes that affect the migration of the intercortical neurons during embryonic development (Valiente and Marin, 2010). In addition, for decades we have known that there is a correlation between schizophrenia and the alteration of visual perception and eye movements. In fact, just recently it has been shown that simple tests for the detection of abnormal eye movements can discriminate cases of schizophrenia from controls with exceptional accuracy (Benson et al., 2012). Of the genes that are involved in the disrupted tangential migration of cortical neurons, NRG1, ERBB4, GRIN1, DISC1, and DTNBP1 (Marin, 2012), four of them are involved in the expression of molecules related with visual structures, and one of them is related to early visual processing: NRG1 is expressed in the cornea (Brown et al., 2004) and one of its mutations (rs3924999) affects spatial accuracy on the anti-saccade (AS) task (Schmechtig et al., 2010) and is associated with auditory P300 in schizophrenia (Kang et al., 2012);

Projection transform on spatio-temporal context for action recognition

This paper discusses the task of human action recognition. This task is important to applications like video surveillance and video retrieval. Most of the existing local interest points based works on human action analysis, lost the information about spatio-temporal distribution of features and neglected the relationship between features and each defined actions. In this paper, through the analysis of feature distribution and their interactions over spatio-temporal domain, we propose a novel projection transform to take the two factors into account. A video sequence of human action in our perspective can be modeled by three types of features of spatio-temporal interest points: the global projection transform feature, the relative position distribution feature and the bag of visual words based feature. Then a new context-K-nearest-neighbor classifier is utilized to fuse them to form discriminative feature sets for action matching. In most of the case, our experiments have indicated that the novel method outperforms other previous published results on the Weizmann and KTH datasets.

Iterative Human Pose Estimation based on A New Part Appearance Model

Human pose estimation has become a hot topic in the field of computer vision , it can be used in human activity analysis, video, and any other field ,it main purpose is that detect the position, scale and direction of parts of people .Because of the result in the iterative human pose estimation based on tree-structure model is susceptible to the background, In this paper(consider static images),part appearance model is improved, an appearance model based on colour and texture for iterative human pose estimation is proposed, experimental results show this method give a better performance and accuracy while reduce the search space by people detector and grab cut.

3D human motion analysis framework for shape similarity and retrieval

3D shape similarity from video is a challenging problem lying at the heart of many primary research areas in computer graphics and computer vision applications. In this paper, we address within a new framework the problem of 3D shape representation and shape similarity in human video sequences. Our shape representation is formulated using extremal human curve (EHC) descriptor extracted from the body surface. It allows taking benefits from Riemannian geometry in the open curve shape space and therefore computing statistics on it. It also allows subject pose comparison regardless of geometrical transformations and elastic surface change. Shape similarity is performed by an efficient method which takes advantage of a compact EHC representation in open curve shape space and an elastic distance measure. Thanks to these main assets, several important exploitations of the human action analysis are performed: shape similarity computation, video sequence comparison, video segmentation, video clustering, summarization and motion retrieval.Experiments on both synthetic and real 3D human video sequences show that our approach provides an accurate static and temporal shape similarity for pose retrieval in video, compared with the state-of-the-art approaches. Moreover, local 3D video retrieval is performed using motion segmentation and dynamic time warping (DTW) algorithm in the feature vector space. The obtained results are promising and show the potential of this approach.

Pain Expression Recognition Based on Fuzzy Classification Approach

Human activity analysis and expressions recognition play a central role in pervasive health care. Expressions recognition is an active research field in computer vision. In this paper, We propose an approach for automatically recognizing the pain expression from video sequences, in which pain is categorized into four levels, including “no-pain”, “slight pain”, “moderate pain” and “severe pain”. First of all, each frame in video sequences is represented as a “word” by extracting facial velocity information, which is used to charaterize pain expression. Then boosting fuzzy classification algorithm is used for pain expression recognition. The algorithm can automatically categorize and recognize the human pain expression which is contained in the video. Finally we test our algorithm on a pain expression dataset built by ourselves. Experiment results show that the average recognition accuracy is over 91%, which validates its effectiveness. Our results are significantly better than three state-of-the-art approaches for pain expression recognition on the same data.

Calibration of D-RGB camera networks by skeleton-based viewpoint invariance transformation

Combining depth information and color image, D-RGB cameras provide a ready detection of human and associated 3D skeleton joints data, facilitating, if not revolutionizing, conventional image centric researches in, among others, computer vision, surveillance, and human activity analysis. Applicability of a D-RBG camera, however, is restricted by its limited range of frustum of depth in the range of 0.8 to 4 meters. Although a D-RGB camera network, constructed by deployment of several D-RGB cameras at various locations, could extend the range of coverage, it requires precise localization of the camera network: relative location and orientation of neighboring cameras. By introducing a skeleton-based viewpoint invariant transformation (SVIT), which derives the relative location and orientation of a detected humans upper torso to a D-RGB camera, this paper presents a reliable automatic localization technique without the need for additional instrument or human intervention. By respectively applying SVIT to two neighboring D-RGB cameras on a commonly observed skeleton, the respective relative position and orientation of the detected humans skeleton for these two cameras can be obtained before being combined to yield the relative position and orientation of these two cameras, thus solving the localization problem. Experiments have been conducted in which two Kinects are situated with bearing differences of about 45 degrees and 90 degrees; the coverage can be extended by up to 70% with the installment of an additional Kinect. The same localization technique can be applied repeatedly to a larger number of D-RGB cameras, thus extending the applicability of D-RGB cameras to camera networks in making human behavior analysis and context-aware service in a larger surveillance area.

A Quantitative Analysis of the Human Resources Activity of the Iranian Cyber Army

The Iranian cyberspace has been developed in a natural process of Internet technologies being introduced in the society for more than a decade. However, governmental policies have had dominant impacts on its development. Specifically, during the campaigns for presidential elections in 2009, in the absence of freedom of speech in real world and stagnancy of traditional media in providing a criticising content in favour of democracy, citizen journalism played a major role in covering the news in the independent space of the web. Since then, the Iranian state authorities have inevitably decided for new strategic plans in order to control and object the flow of free information through virtual social networks. One of the strategies inducing publicly that cyberspace and its provided opportunities are threats to national security and therefore, establishing and enhancing a cyber army to save the virtual battlefield have been adopted as an implicative response to the situation. Now, it is quite arguable to claim that Iranian cyber army has operated successfully and the human resources who support the governmental policies and could be referred as pro-government warriors in soft war are still ready to keep on their predefined roles. This research activity, including an initial survey on the literature of soft war and a brief chronology of formation of the Iranian cyber army, performs an analysis on a number of determined statistical features of the online activity of these pro-government warriors in order to evaluate the validity of the acclaimed statement. In this paper, by referring to the quantitative outcomes of the study, it is tried to clarify whether the Iranian state authorities have successfully managed pro-government cyber warriors or not.

Detection of Unusual Human Activities Based on Behavior Modeling

A type of services that require human physical actions and intelligent decision making exists in various real fields, such as nursing in hospitals and caregiving in nursing homes. In this paper, we propose new formalism for modeling human behavior in such services. Behavior models are estimated from event-logs, and can be used for analysis of human activities. We show two analysis methods: one is to detect unusual human activities that appear in event-logs, and the other is to find staffs that behave differently from others.

An Improved Structural Human Pose Model

The human pose model is the key technology of the human-computer interaction and action analysis and understanding. In this paper, an improved structural human pose model is proposed, it integrated the capability of the structural analysis in the picture structural model and the advantage of the free view of the simple staff model, with the attribute of the structure of the model, the capability of the grammar analysis is boosted up, it is the base of the human action analysis and understanding.

Space-time Facet Model for Human Activity Classification

This paper presents a novel space-time feature-based human activity analysis system. We detect Space Time Interest Points (STIP) and generate their description based on the facet model. The proposed approach detects interest points in video data using the three-dimensional facet model efficiently. Then we describe each interest point by three-dimensional Haar wavelet transform and time derivatives of different order obtained from said facet model. Here we represent each video clip following the bag-of-words approach by learning feature specific dictionary. Finally, classification is done using non-linear SVM with χ 2 -kernel. We evaluate the performance of our system on standard datasets like Weizmann, KTH, UCF sports, ICD, UCF YouTube, and UCF50 and get better, or at least comparable results compared to other state-of-the-art systems.