Features Of Human Activities Research Articles

Using a Multilearner to Fuse Multimodal Features for Human Action Recognition

The representation and selection of action features directly affect the recognition effect of human action recognition methods. Single feature is often affected by human appearance, environment, camera settings, and other factors. Aiming at the problem that the existing multimodal feature fusion methods cannot effectively measure the contribution of different features, this paper proposed a human action recognition method based on RGB-D image features, which makes full use of the multimodal information provided by RGB-D sensors to extract effective human action features. In this paper, three kinds of human action features with different modal information are proposed: RGB-HOG feature based on RGB image information, which has good geometric scale invariance; D-STIP feature based on depth image, which maintains the dynamic characteristics of human motion and has local invariance; and S-JRPF feature-based skeleton information, which has good ability to describe motion space structure. At the same time, multiple K-nearest neighbor classifiers with better generalization ability are used to integrate decision-making classification. The experimental results show that the algorithm achieves ideal recognition results on the public G3D and CAD60 datasets.

S3D-CNN: skeleton-based 3D consecutive-low-pooling neural network for fall detection

Most existing deep-learning-based fall detection methods use either 2D neural network without considering movement representation sequences, or whole sequences instead of only those in the fall period. These characteristics result in inaccurate extraction of human action features and failure to detect falls due to background interferences or activity representation beyond the fall period. To alleviate these problems, a skeleton-based 3D consecutive-low-pooling neural network (S3D-CNN) for fall detection is proposed in this paper. In the S3D-CNN, an activity feature clustering selector is designed to extract the skeleton representation in depth videos using pose estimation algorithm and form optimized skeleton sequence of fall period. A 3D consecutive-low-pooling (3D-CLP) neural network is proposed to process these representation sequences by improving network in terms of layer number, pooling kernel size, and single input frame number. The proposed method is evaluated on public and self-collected datasets respectively, outperforming the existing methods.

How social contexts affect cognition: Mentalizing interferes with sense of agency during voluntary action

Living in complex social structures, humans have evolved a unique aptitude for mentalizing: trying to understand and predict the behaviour of others. To date, little is known about how mentalizing interacts with other cognitive processes. “Sense of agency” refers to the feeling of control over the outcomes of one's actions, providing a precursor of responsibility. Here, we test a model of how social context influences this key feature of human action, even when action outcomes are not specifically social. We propose that in social contexts, sense of agency is affected by the requirement to mentalize, increasing the complexity of individual decision-making. We test this hypothesis by comparing two situations, in which participants could either consider potential actions of another person (another participant acting to influence the task), or potential failures of a causal mechanism (a mechanical device breaking down and thereby influencing the task). For relatively good outcomes, we find an agency-reducing effect of external influence only in the social condition, suggesting that the presence of another intentional agent has a unique influence on the cognitive processes underlying one's own voluntary action. In a second experiment, we show that the presence of another potential agent reduces sense of agency both in a context of varying financial gains or of losses. This clearly dissociates social modulation of sense of agency from classical self-serving bias. Previous work primarily focused on social facilitation of human cognition. However, when people must incorporate potential actions of others into their decision-making, we show that the resulting socio-cognitive processes reduce the individuals' feelings of control.

Human action recognition using Lie Group features and convolutional neural networks

In recent years, skeleton-based human action recognition has attracted substantial attentions. However, owing to the complexity and nonlinearity of human action data, it is still a challenging task to precisely represent skeleton features. Motivated by the effectiveness of Lie Group skeletal representations in extracting human action features and the powerful capability of deep neural networks in feature learning and high-dimensional data processing, we proposed to combine Lie Group features and deep learning for human action recognition. Human skeleton information was firstly used to overcome the interference of external factors such as changes of lighting conditions and body shape. And then, Lie Group was applied to naturally represent the complex and diverse action data. Finally, we took use of convolutional neural networks to learn and classify the Lie Group features. Experiments were performed on three public datasets, and the experimental results show that our methods can achieve higher average recognition accuracy of 93.00% on Florence3D-Action, 93.68% on MSR Action Pairs, and 97.96% on UT Kinect-Action, which outperforms many of the state-of-the-art methods.

Fall Detection in Videos With Trajectory-Weighted Deep-Convolutional Rank-Pooling Descriptor

Automatic fall detection in videos could enable timely delivery of medical service to the injured elders who have fallen and live alone. Deep ConvNets have been used to detect fall actions. However, there still remain problems in deep video representations for fall detection. First, video frames are directly inputted to deep ConvNets. The visual features of human actions may be interfered with surrounding environments. Second, redundant frames increase the difficulty of time encoding for human actions. To address these problems, this paper presents trajectory-weighted deep-convolutional rank-pooling descriptor (TDRD) for fall detection, which is robust to surrounding environments and can describe the dynamics of human actions in long time videos effectively. First, CNN feature map of each frame is extracted through a deep ConvNet. Then, we present a new kind of trajectory attention map which is built with improved dense trajectories to optimally localize the subject area. Next, the CNN feature map of each frame is weighted with its corresponding trajectory attention map to get trajectory-weighted convolutional visual feature of human region. Further, we propose a cluster pooling method to reduce the redundancy of the trajectory-weighted convolutional features of a video in the time sequence. Finally, rank pooling method is used to encode the dynamic of the cluster-pooled sequence to get our TDRD. With TDRD, we get superior result on SDUFall dataset and get comparable performances on UR dataset and Multiple cameras dataset with SVM classifiers.

Michel Foucault and the Problematics of Power: Theorizing DTCA and Medicalized Subjectivity.

This article explores Foucault's two different notions of power: one where the subject is constituted by power-knowledge relations and another that emphasizes how power is a central feature of human action. By drawing out these two conceptualizations of power, Foucault's work contributes three critical points to the formation of medicalized subjectivities: (1) the issue of medicalization needs to be discussed both in terms of both specific practices and holistically (within the carceral archipelago); (2) we need to think how we as human beings are "disciplined" and "subjectivated" through medicalization, as discourses, practices, and institutions are all crystallizations of power relations; and (3) we need to reflect on how we can "resist" this process of subjectification, since "power comes from below" and patients shape themselves through "technologies of the self." Ultimately, Foucault's work does not merely assist us in refining our analysis; rather, it is essential for conceptualizing medicalization in contemporary society.

Human Action Recognition Using Improved Sparse Gaussian Process Latent Variable Model and Hidden Conditional Random Filed

Owing to the fragile to external environment, current human activity recognition is still challenging in the area of computer vision. In particular, the recognition of subtle actions is a difficult task to distinguish from each other. In this paper, we proposed the methods for human action recognition based on improved sparse Gaussian process latent variable model (GPLVM) and hidden conditional random field (HCRF). We first fused the skeletal information and the motion characteristics of human body to extract human action features from a set of action sequences. And then, we proposed an improved sparse GPLVM algorithm for feature dimensionality reduction to satisfy the visualization and computation complexity for human actions recognition. Through sparse approximation with information vector machine and the dynamic process of human actions, the improved sparse GPLVM allows the action features to be represented by the low dimensional trajectory in manifold space. Furthermore, HCRF was applied to recognize the characteristics of human actions derived from manifold spaces and finally to classify human actions. The proposed methods are tested and evaluated on public action database. The experimental results show that our methods can achieve better performance of feature dimensionality reduction and visualization and obtain the average recognition accuracy of 93.68%.

Human Action Recognition Algorithm Based on Adaptive Initialization of Deep Learning Model Parameters and Support Vector Machine

Action recognition is a trending topic and key research direction in computer vision, machine learning, artificial intelligence, and other fields. This research seeks to identify human action in image and video data. Its research results have been widely used in the fields of safety monitoring, disability monitoring, understanding multimedia content, human–computer interaction, virtual reality, and so on. However, the existing traditional human action recognition technology has many limitations in practical application, such as low accuracy and weak adaptive ability. Although the action recognition based on deep learning can self-learn and improve the action recognition accuracy, there are many difficulties in training the deep neural network model, such as gradient disappearance, gradient explosion, and overfitting. Therefore, this paper will reduce the abovementioned difficulties in deep neural network model training from the perspective of deep neural network model parameter initialization and then propose a model parameter initialization method based on the multilayer maxout network activation function to solve the difficulties in deep neural network model training. Then, on this basis, a method of learning the temporal and spatial characteristics of human action based on the deep neural network model is proposed. First, the method detects and tracks the human action and uses the restricted Boltzmann machine (RBM) to encode the temporal and spatial features of various parts of the human body. Second, the temporal and spatial feature codes of various parts of the human body are integrated into a global temporal and spatial feature representation method of the action video through a RBM neural network. Finally, the trained SVM classifiers are used to recognize human action. Experiments show that the human action recognition method proposed in this paper not only has high recognition accuracy but also has great adaptability. Thus, this method extracts temporal and spatial features from the shape feature sequences of various parts of the human body, thus opening up a new way to extract human action features and solving the problem of human action recognition in complex scenes. Its proposal provides an exploratory technical method and approach for self-adaptive recognition of human action. It also gives directional enlightenment to the development and improvement of self-adaptive human action methods.

Recognizing human actions from silhouettes described with weighted distance metric and kinematics

A virtual particle random walking theory under variable velocities is presented in this paper. Under the proposed theory, the solutions of some two-dimensional Poisson equations, which are discretized by nine-point finite difference method and defined on the so-called spatial-temporal motion accumulative image stemming from human silhouettes, provide us the depth contour image for actions description. Although merely two-dimensional definition domain and concepts are related to the Poisson equations, both spatial and temporal evolution information of human actions are successfully included in the depth contour image owing to designating the travelling velocities of virtual particles according to the spatial-temporal motion accumulative image. In addition, it is worth noting that projecting three-dimensional human actions to the two-dimensional image descriptors contributes to much lower computation cost in the corresponding recognition algorithms, compared to those when using the three-dimensional spatial-temporal descriptors directly. In order to enhance the recognition accuracy, a hierarchical cascaded classifier is configured with cascading nearest neighbor classifiers, in each layer of which different kinds of shape and kinematic features of human actions are dealt with. Numerical experimental results on several public human action databases are illustrated to verify recognition performance improvements by means of the proposed algorithm.

3D Human Action Recognition using Hu Moment Invariants and Euclidean Distance Classifier

This paper presents a new model of scale, rotation, and translations invariant interest point descriptor for human actions recognition. The descriptor, HMIV (Hu Moment Invariants on Videos) is used for solving surveillance camera recording problems under different conditions of side, position, direction and illumination. The proposed approach deals with raw input human action video sequences. Seven Hu moments are computed for extracting human action features and for storing them in a 1D vector which is constringed as one mean value for all the frames’ moments. The moments are invariant to scale, translation, or rotation, which is the robustness point of Hu moments algorithm. The experiments are evaluated using two different datasets; KTH and UCF101. The classification process is executed by calculating the Euclidean distance between the training and testing datasets. Human action with minimum distance will be selected as the winner matching action. The maximum classification accuracy in this work is 93.4% for KTH dataset and 92.11% for UCF101.

Quantification of performance shaping factors (PSFs)’ weightings for human reliability analysis (HRA) of low power and shutdown (LPSD) operations

Recently, human actions during low power and shutdown (LPSD) operation has been recognized as important contributors to the safe operation of nuclear power plants (NPPs). The features of human actions differ between LPSD operation and full power operation because automatic control may be disabled, most control room annunciation tiles indicate alarmed status, and procedures are prepared with less than sufficiency. The goal of this study is to quantify the weightings of performance shaping factors (PSFs) when performing HRA during LPSD operation. In HRA, PSFs are the conditions under which an operator performs the tasks, and PSFs can increase or decrease the probability of human error. In this study, in order to quantify the weightings of PSFs, a profiling technique was adopted. To this end, human error data were collected from domestic NPP operational experience and the PSFs were investigated for describing each human error datum by using a root cause analysis method. As a result, the probabilities of human errors were provided and the weightings of several single PSF and combined PSFs which affect the probabilities of human errors were quantified when implementing HRA during LPSD operation.

Esquema de una teoría del agente centrada en la creatividad

Las diferentes tradiciones teóricas de la sociología han reconocido que la inventiva, la imaginación y la creatividad son rasgos inherentes a la acción humana. Sin embargo, han privilegiado en sus teorizaciones otros aspectos de la acción, como la racionalidad, el seguimiento de normas sociales, la espontaneidad de las prácticas cotidianas o las habilidades lingüísticas y comunicativas. La intención de este artículo es proponer un esquema amplio y flexible para el desarrollo de una concepción del agente centrada en la creatividad, cuyo interés para las ciencias sociales viene dado por el vínculo, intuitivamente obvio, pero extremadamente complejo, entre creatividad y emergencia de nuevas identidades y nuevas formas de organización social. El esquema parte de una concepción del agente como estratificación de niveles de consciencia y autoconocimiento, diferenciando un plano pulsional de la creatividad, un plano prereflexivo-práctico y un plano discursivo. En cada uno se identifican mecanismos específicos de la creatividad de la acción, para esbozar luego algunas hipótesis de articulación entre los tres y algunas líneas de investigación que podrían, a futuro, apuntalar la propuesta.

Women in a multicultural diaspora: dilemmas of gender and culture

Redefining culture as ‘a dynamic and evolving feature of human action', feminists stress that cultures are not static, establishing the constant need to interrogate them, both in innate and distant contexts, namely when women are placed in their immediate cultural contexts or when they migrate to a new setting carrying their cultures along, wherein both events entail multiple expressions and interpretations. For instance, how is one to juxtapose instances of honour killing, dowry abuses, forced marriages, female genital mutilation with women being able to ‘transgress’ home barriers in the new (outside, foreign and external) locations, ably diluting, if not completely dismantling, the extremities of their ascribed cultural systems, and sometimes even joining at high-level posts at multi national companies (MNCs) and so on. The ‘new freedom’ from cultural specificities is attributed to the general openness and proclamations of liberty that abound the new multicultural contexts with the parallel process of a snowball-like cultural regress. Women in Diaspora then throw up the simultaneous issues of discrimination and empowerment, emerging from their distancing from and sometimes forced clinching to their regressive cultures. This gathers two responses within feminism vis-à-vis multiculturalism. The homogenized and hierarchical defining of cultures necessitates unbiased interrogation to identify the real beneficiary, for women's life – choices are not always thrust upon them by patriarchal and suppressive cultures and Diaspora offers women liberatory avenues too, despite certain crippling cultural tendencies tainting it.

The Aristotelian conception of habit and its contribution to human neuroscience.

The notion of habit used in neuroscience is an inheritance from a particular theoretical origin, whose main source is William James. Thus, habits have been characterized as rigid, automatic, unconscious, and opposed to goal-directed actions. This analysis leaves unexplained several aspects of human behavior and cognition where habits are of great importance. We intend to demonstrate the utility that another philosophical conception of habit, the Aristotelian, may have for neuroscientific research. We first summarize the current notion of habit in neuroscience, its philosophical inspiration and the problems that arise from it, mostly centered on the sharp distinction between goal-directed actions and habitual behavior. We then introduce the Aristotelian view and we compare it with that of William James. For Aristotle, a habit is an acquired disposition to perform certain types of action. If this disposition involves an enhanced cognitive control of actions, it can be considered a “habit-as-learning”. The current view of habit in neuroscience, which lacks cognitive control and we term “habit-as-routine”, is also covered by the Aristotelian conception. He classifies habits into three categories: (1) theoretical, or the retention of learning understood as “knowing that x is so”; (2) behavioral, through which the agent achieves a rational control of emotion-permeated behavior (“knowing how to behave”); and (3) technical or learned skills (“knowing how to make or to do”). Finally, we propose new areas of research where this “novel” conception of habit could serve as a framework concept, from the cognitive enrichment of actions to the role of habits in pathological conditions. In all, this contribution may shed light on the understanding of habits as an important feature of human action. Habits, viewed as a cognitive enrichment of behavior, are a crucial resource for understanding human learning and behavioral plasticity.

Proneness for exercise, cognitive and psychophysiological consequences of action observation

ObjectivesPhysical exercise has benefits that go beyond health and well-being, namely in cognitive, motor and psychophysiological areas. The discovery of a shared neural network between action observation and execution (Action-Observation Network) led us to hypothesize that watching human motor action might improve cognitive and motor aspects of performance and proneness for exercise. Design/MethodsSixty participants viewed a Motor (M) (n = 30) or a Non-Motor (NM) (n = 30) movie with strong or weak content of motoric features of human action, respectively. Performance in d2 Attention test, Fitts' Motor task, and a Proneness for Exercise Visual Analog Scale was assessed before and after movie visualization, in a cross-sectional study. Psychophysiological measures were recorded throughout the experiment. ResultsOur results demonstrate an increase in proneness for exercise, and greater improvement in attention-related cognitive aspects in the M group. The aforementioned benefits of action observation did not modulate motor performance. A mental effort deployment was associated to the decrease in heart rate variability after visualization of the NM movie. This was not conducive to attention channeling on task performance. Conversely, M movie observation seemed to be associated to a cognitive load release, affording attention deployment for the resolution of the subsequent tasks. ConclusionsIt seems that some benefits associated to physical practice can result from the mere visualization of movies with human motor action content. These are the improvement in attention-related cognitive skills associated to psychophysiological changes that support a disengagement from mental effort. Crucially, the observation of exercise behavior seems to be a key factor for exercise adherence.

‘Could you define the sense you give the word “political”’? Michel Foucault as a political philosopher

Foucault’s political thinking is focused on the concept of power relations. Under the influence of Nietzsche he proposes two different accounts of how power is related to human action. Nietzsche had argued, on the basis of a reading of Kant’s antinomies of pure reason, for two different accounts of that relationship. On the one hand, he had sought to understand action as a phenomenon of the will to power; on the other, he had also spoken of the will to power as an aspect of human agency. On Nietzsche’s views, we need to assert both positions even though they are for us irreconcilable. In his writings on power and action Foucault finds himself driven into adopting a similarly dual view. While he speaks of action in the nineteen seventies as subsidiary to power relations, he reverses himself in the nineteen eighties by treating power as a feature of human action. Just like Nietzsche, he offers us no way of reconciling these two distinct accounts.

Enough skill to kill: Intentionality judgments and the moral valence of action

Extant models of moral judgment assume that an action’s intentionality precedes assignments of blame. Knobe (2003b) challenged this fundamental order and proposed instead that the badness or blameworthiness of an action directs (and thus unduly biases) people’s intentionality judgments. His and other researchers’ studies suggested that blameworthy actions are considered intentional even when the agent lacks skill (e.g., killing somebody with a lucky shot) whereas equivalent neutral actions are not (e.g., luckily hitting a bull’s-eye). The present five studies offer an alternative account of these provocative findings. We suggest that people see the morally significant action examined in previous studies (killing) as accomplished by a basic action (pressing the trigger) for which an unskilled agent still has sufficient skill. Studies 1 through 3 show that when this basic action is performed unskillfully or is absent, people are far less likely to view the killing as intentional, demonstrating that intentionality judgments, even about immoral actions, are guided by skill information. Studies 4 and 5 further show that a neutral action such as hitting the bull’s-eye is more difficult than killing and that difficult actions are less often judged intentional. When difficulty is held constant, people’s intentionality judgments are fully responsive to skill information regardless of moral valence. The present studies thus speak against the hypothesis of a moral evaluation bias in intentionality judgments and instead document people’s sensitivity to subtle features of human action.

Causation and aggression

F or libertarians, the purpose of a legal system is to establish and enforce rules that facilitate and support peaceful, conflict-free interaction between individuals. In short, the law should prohibit aggression. Because aggression is a particular kind of human action--action that intentionally violates or threatens to violate the physical integrity of another person or another person's property without that person's consent--it can be successfully prohibited only if the law is based on a sound understanding of the nature of human action more generally. Praxeology, the general theory of human action, studies the universal features of human action and draws out the logical implications of the undeniable fact that humans act (Mises 1966, pp. 15-16, 480; and 1978; Hoppe 1995). Praxeology is central to Austrian economics, the hitherto best elaborated part of the science of praxeology (Mises 1966, p. 3). However, other disciplines can benefit from the insights of praxeology. Hans-Hermann Hoppe has already extended praxeology to the field of political ethics (Hoppe 1989b, chap. 7). The related discipline of legal theory, which also concerns ethical implications of human action, can also benefit from the insights of praxeology. In the context of legal analysis, one important praxeological doctrine is the distinction between action and mere behavior. The difference between action and behavior boils down to intent. Action is an individual's intentional intervention in the physical world, via certain selected means, with the purpose of attaining a state of affairs that is preferable to the conditions that would prevail in the absence of the action. Mere behavior, by contrast, is a person's physical movements that are not undertaken intentionally and that do not manifest any purpose, plan, or design. Mere behavior cannot be aggression; aggression must be deliberate, it must be an action.

Humanity asHomo docens: the Teaching Species

Abstract Teaching, instruction, pedagogy, showing other people how to do things – these are seemingly universal features of human action; yet they do not figure in conventional descriptions of 'human nature'; nor do they appear as prominent items in anthropologists' accounts of simple societies. It is appropriate to mark the tercentenary of John Locke's 'Some thoughts concerning education' by asking: what does this neglect signify?

Building a user model implicitly from a cooperative advisory dialog

This paper reviews existing methods for building user models to support adaptive, interactive systems, identifies sigificant problems with these approaches, and describes a new method for implicitly acquiring user models from an ongoing user-system dialog. Existing explicit user model acquisition methods, such as user edited models or model building dialogs put additional burden on the user and introduce artificial model acquisition dialogs. Hand coding stereotypes, another explicit acquisition method, is a tedious and error-prone process. On the other hand, implicit acquisition techniques such as computing presuppositions or entailments either draw too few inferences to be generally useful, or too many to be trusted. In contrast, this paper describes GUMAC, a General User Model Acquisition Component that uses heuristic rules to make default inferences about users' beliefs from their interaction with an advisory expert system. These rules are based on features of human action and conversation that constrain people's behavior and establish expectations about their knowledge. The application of these rules is illustrated with two examples of extended dialogs between users and an investment advisory system. During the course of these conversations, GUMAC is able to acquire an extensive model of the users' beliefs about the aspects of the domain considered in the dialog. These models, in turn, provide the sort of information needed by an explanation generator to tailor explanations the advisory system gives to its users.