Human Activity Analysis Research Articles

Drone-Action: An Outdoor Recorded Drone Video Dataset for Action Recognition

Aerial human action recognition is an emerging topic in drone applications. Commercial drone platforms capable of detecting basic human actions such as hand gestures have been developed. However, a limited number of aerial video datasets are available to support increased research into aerial human action analysis. Most of the datasets are confined to indoor scenes or object tracking and many outdoor datasets do not have sufficient human body details to apply state-of-the-art machine learning techniques. To fill this gap and enable research in wider application areas, we present an action recognition dataset recorded in an outdoor setting. A free flying drone was used to record 13 dynamic human actions. The dataset contains 240 high-definition video clips consisting of 66,919 frames. All of the videos were recorded from low-altitude and at low speed to capture the maximum human pose details with relatively high resolution. This dataset should be useful to many research areas, including action recognition, surveillance, situational awareness, and gait analysis. To test the dataset, we evaluated the dataset with a pose-based convolutional neural network (P-CNN) and high-level pose feature (HLPF) descriptors. The overall baseline action recognition accuracy calculated using P-CNN was 75.92%.

Quantifying human error probability in independent protection layers for a batch reactor system using dynamic simulations

The batch processing facilities with features involving in large-scale human intervention are generally considered to be relatively hazardous processes due to frequent change in operating conditions. Therefore, sufficient protection layers should be implemented against the hazards inherent in the batch processing facilities, especially for batch reactor systems. The human operations involved in the batch process play an important role to maintain process production and also take responsibility for reducing risks in specific possible accident scenarios. However, in practice, a number of companies initially did not take any credits for human actions in Layer of Protection Analysis (LOPA) or were overly optimistic in a few cases, which result in unwarranted safety expenditures or inadequate risk reductions. In this study, we propose a simulation-oriented methodology for quantifying human error probability (HEP) in independent protection layers (IPLs). The methodology integrates dynamic simulations with human reliability analysis (HRA) to give a comprehensive analysis of human actions. Case study for a batch reactor system shows that risk reductions of human response are expected to be greater than that determined using other methods. The present study provides a solution for analyzing the risk reductions of human elements within the IPL and serves as a reference for improving the emergency response for operators.

Improvements of the Analysis of Human Activity Using Acceleration Record of Electrocardiographs

The use of Holter Electrocardiograph (Holter ECG) is rapidly spreading. It is a wearable electrocardiograph that records 24-hour electrocardiograms in a built-in flash memory, making it possible to detect atrial fibrillation (Atrial Fibrillation, AF) through all-day activities. It is also useful for screening for diseases other than atrial fibrillation and for improving health. It is said that more useful information can be obtained by combining electrocardiograph with the analysis of physical activity. For that purpose, the Holter electrocardiograph is equipped with heart rate sensor and acceleration sensors. If acceleration data is analysed, we can estimate activities in daily life, such as getting up, eating, walking, using transportation, and sitting. In combination with such activity status, electrocardiographic data can be expected to be more useful. In this study, we investigate the estimation of physical activity. For the better analysis, we evaluated activity estimation using machine learning as well as several different feature extractions. In this report, we will show several different feature extraction methods and result of human body analysis using machine learning.

ЭЛЕМЕНТЫ СОЦИАЛЬНОЙ МАРКИРОВАННОСТИ В РОССИЙСКОМ ОФИЦИОЗНОМ ДИСКУРСЕ

Statement of the research problem. One of the most urgent problems of speech communication, despite the deep research in this area, continues to be the adequacy of the understanding of the statement addressed from the addresser to the addressee. This situation is explained not only by the social nature of the language and the variety of situations of its use, but also by frequent cases of either unsatisfactory verbal embodiment of the speaker’s intentions, or insufficient preparation of the listener for an unmistakable perception of the meaning of the verbal act intended for him. The most tangible is the resonance produced in society by public statements of representatives of the hierarchically dominant stratum. The purpose of this article is discourse-analysis of statements (texts) of officials who have received a wide coverage in the Internet environment, which is to identify explicit and implicit markers of semantics, style and pragmatics of statements of officials. Methodology. The basic foundation of the research methodology is the works by M.M. Bakhtin, Yu.M. Lotman, E.D. Polivanov, A.A. Potebnya, L.V. Shcherba, T. van Dijk and other researchers who laid the foundations for understanding human communicative activity and discursive analysis practices. The novelty of the study is that as a result of the analysis of statements by official government representatives the authors found the root cause of “failures” in speech situations of semiofficial discourse. Conclusion. The reasons for the described situations of mutual misunderstanding is the lack of general social experience of speakers and perceivers of the text. The author’s contribution lies in the discursive analysis of the statements which caused the dissonance and the identification of explicitly and implicitly expressed markers of social marking of these texts.

A Hierarchical Model for Human Action Recognition From Body-Parts

As increasing attention is paid to human action recognition from skeleton data, this paper focuses on such tasks by proposing a hierarchical model to discover the structure information of body-parts involved in actions for better analysis of human actions in the skeleton data. Considering human actions as simultaneous motions of body-parts of the human skeleton, we propose a hierarchical model to simultaneously apply discriminative body-parts selection at a same scale and group coupling of bundles of body-parts at different scales, while we decompose the human skeleton into a hierarchy of body-parts of varying scales. To represent such hierarchy of body-parts, we accordingly build a hierarchical rotation and relative velocity (HRRV) descriptor. The hierarchical representations encoded by Fisher vectors of the HRRV descriptors are properly formulated into the hierarchical model via the proposed mixed norm, to apply the sparse selection of body-parts and regularize the structure of such hierarchy of body-parts. The extensive evaluations on three challenging datasets demonstrate the effectiveness of our proposed approach, which achieves superior performance compared to the state-of-the-art algorithms on datasets with various sizes, showing it is more widely applicable than existing approaches.

RECOGNITION OF URBAN FUNCTIONAL REGIONS AT STREET SCALE BASED ON LJ1-01 NIGHT-TIME LIGHT REMOTE SENSING AND MOBILE BIG DATA

Abstract. Night-time light (NTL) remote sensing data has been widely used in the analysis of human activities at night for a large extent, such as light pollution monitoring and recognition of urban functional regions. In most previous studies, the spatial resolutions of NTL remote sensing data are rather low (e.g., 500 m or coarser). Besides, remote sensing classification of land use rather than land cover types is often a hard task. Due to the reasons, it is difficult to meet the demand of urban refined management. In this study, mobile big data and Luojia1-01 (LJ1-01) NTL remote sensing satellite data with higher spatial resolution are adopted for extracting urban functional regions at the street-level scale. Taking Shenzhen city as a case, mobile big data (i.e., the volume of mobile devices) with the accuracy of approximate 150 m (i.e., 7-bit GeoHash encoding format) is integrated with NTL remote sensing data. After a hot spot analysis, the correlation between the two types of data are analysed. The typical supervised classification algorithms including KNN, SVM and random forest are used to distinguish urban functional regions, which are represented by five types, namely residential, business and commercial, industrial, transportation and other areas. The results show the effectiveness in extracting land use types in cities by combination of Luojia1-01 night-time light remote sensing and mobile big data. This study provides a basis for monitoring night light pollution of residential area, urban planning and the construction of smart cities.

An Unsupervised Framework for Online Spatiotemporal Detection of Activities of Daily Living by Hierarchical Activity Models.

Automatic detection and analysis of human activities captured by various sensors (e.g., sequences of images captured by RGB camera) play an essential role in various research fields in order to understand the semantic content of a captured scene. The main focus of the earlier studies has been widely on supervised classification problem, where a label is assigned to a given short clip. Nevertheless, in real-world scenarios, such as in Activities of Daily Living (ADL), the challenge is to automatically browse long-term (days and weeks) stream of videos to identify segments with semantics corresponding to the model activities and their temporal boundaries. This paper proposes an unsupervised solution to address this problem by generating hierarchical models that combine global trajectory information with local dynamics of the human body. Global information helps in modeling the spatiotemporal evolution of long-term activities, hence, their spatial and temporal localization. Moreover, the local dynamic information incorporates complex local motion patterns of daily activities into the models. Our proposed method is evaluated using realistic datasets captured from observation rooms in hospitals and nursing homes. The experimental data on a variety of monitoring scenarios in hospital settings reveals how this framework can be exploited to provide timely diagnose and medical interventions for cognitive disorders, such as Alzheimer’s disease. The obtained results show that our framework is a promising attempt capable of generating activity models without any supervision.

A Survey of Vision-Based Human Action Evaluation Methods.

The fields of human activity analysis have recently begun to diversify. Many researchers have taken much interest in developing action recognition or action prediction methods. The research on human action evaluation differs by aiming to design computation models and evaluation approaches for automatically assessing the quality of human actions. This line of study has become popular because of its explosively emerging real-world applications, such as physical rehabilitation, assistive living for elderly people, skill training on self-learning platforms, and sports activity scoring. This paper presents a comprehensive survey of approaches and techniques in action evaluation research, including motion detection and preprocessing using skeleton data, handcrafted feature representation methods, and deep learning-based feature representation methods. The benchmark datasets from this research field and some evaluation criteria employed to validate the algorithms’ performance are introduced. Finally, the authors present several promising future directions for further studies.

THE EXPRESSION OF THE CONCEPTUAL OPPOSITION ALLY (PARTNER) – ENEMY (ADVERSARY) IN ENGLISH MILITARY JOURNALISTIC TEXTS

The article is devoted to the ways of representing the concepts of «ally (partner)» and «enemy (adversary)» in modern military-publicistic texts in English. The studied language means have confirmed the existence of these concepts created by the authors of English military genres and the fact that these concepts are fundamental in the military discourse. The study of the conceptual area occurs through the analysis of human cognitive activity. The concept, according to the definition given by Z.D. Popova and I.A. Sternin, is a phenomenon of consciousness, it is modeled according to the signs of its manifestation [Popova, Sternin, 2007, 163]. These signs of the concepts are: culture, habits, education, religion, political relations, values and traditions. The analysis of lexical means objectifying these concepts is carried out on the basis of printed (2006–2007) and electronic journal and newspaper texts (2018–2019). Based on the lexical, semantic, contextual and quantitative analysis of vocabulary selected by the continuous sampling method, the core and periphery of these concepts have been determined. The revealed language means determined the core and periphery of these concepts. The dominant lexical means allow the authors to determine this opposition in English as ally (partner) – enemy (adversary). The texts of three English genres have been analyzed: an analytical article, an interview and a commentary. The conceptual opposition ally (partner) –the enemy (adversary) finds priority use in the genre of an analytical article. In military texts, in accordance with the value reference points, the estimating perception of the opposition occurs, in the authors’ opinion, with positive connotation without an expressive assessment.

A Three-Dimensional Deep Learning Framework for Human Behavior Analysis Using Range-Doppler Time Points

Deep neural networks have shown promise in the radar-based human activity analysis application. Different from existing deep learning models that take either micro-Doppler spectrograms or range profiles as their input, the proposed method can process micromotion signatures in a 3-D way. In this letter, we first transform radar echoes into range-Doppler (RD) time points and then directly process the point sets via a designed 3-D network called the RD PointNet. In fact, our point model is a discrete representation of the motion trajectory. Through this quantitative model, we can use the 3-D network to simultaneously capture human motion profiles and temporal variations. The motion capture simulations and ultrawideband radar measurements show that the proposed framework can achieve superior classification accuracy and noise robustness when compared with image-based methods.

Change Point Detection for Diversely Distributed Stochastic Processes Using a Probabilistic Method

Unpredicted deviations in time series data are called change points. These unexpected changes indicate transitions between states. Change point detection is a valuable technique in modeling to estimate unanticipated property changes underlying time series data. It can be applied in different areas like climate change detection, human activity analysis, medical condition monitoring and speech and image analyses. Supervised and unsupervised techniques are equally used to identify changes in time series. Even though change point detection algorithms have improved considerably in recent years, several undefended challenges exist. Previous work on change point detection was limited to specific areas; therefore, more studies are required to investigate appropriate change point detection techniques applicable to any data distribution to assess the numerical productivity of any stochastic process. This research is primarily focused on the formulation of an innovative methodology for change point detection of diversely distributed stochastic processes using a probabilistic method with variable data structures. Bayesian inference and a likelihood ratio test are used to detect a change point at an unknown time (k). The likelihood of k is determined and used in the likelihood ratio test. Parameter change must be evaluated by critically analyzing the parameters expectations before and after a change point. Real-time data of particulate matter concentrations at different locations were used for numerical verification, due to diverse features, that is, environment, population densities and transportation vehicle densities. Therefore, this study provides an understanding of how well this recommended model could perform for different data structures.

An adaptive and on-line IMU-based locomotion activity classification method using a triplet Markov model

Detecting locomotion activities are critical for the analysis of human daily activities. In this paper, an adaptive on-line classification method is proposed to detect four lower limb locomotion activities – walking, running, stair ascent and descent – from the signals of a unique wearable sensor. The method is based on a non-parametric triplet Markov model, to detect gait phases and activities simultaneously, in an unsupervised way. This capability allows the model to work at run-time, and so to be used on-line. Also, an algorithm that adapts model parameters suits for a wide range of healthy human is presented. From this adjustment ability, an initial model can gradually approach to the dedicated activity patterns. Experimental results with ten healthy subjects show that our algorithm can reach an overall classification accuracy up to 99.20%, after the stabilization of parameters adjustment, regardless of the users’ gender, height, activity speed… Overall, the proposed algorithm presents a good performance in on-line parameters learning and high accuracy in classifying lower limb locomotion activities from a fount-mounted inertial measurement unit-based wearable sensor.

Fast Graph Fourier Transforms Based on Graph Symmetry and Bipartition

The graph Fourier transform (GFT) is an important tool for graph signal processing, with applications ranging from graph-based image processing to spectral clustering. However, unlike the discrete Fourier transform, the GFT typically does not have a fast algorithm. In this work, we develop new approaches to accelerate the GFT computation. In particular, we show that Haar units (Givens rotations with angle π/4) can be used to reduce GFT computation cost when the graph is bipartite or satisfies certain symmetry properties based on node pairing. We also propose a graph decomposition method based on graph topological symmetry, which allows us to identify and exploit butterfly structures in stages. This method is particularly useful for graphs that are nearly regular or have some specific structures, e.g., line graphs, cycle graphs, grid graphs, and human skeletal graphs. Though butterfly stages based on graph topological symmetry cannot be used for general graphs, they are useful in applications, including video compression and human action analysis, where symmetric graphs, such as symmetric line graphs and human skeletal graphs, are used. Our proposed fast GFT implementations are shown to reduce computation costs significantly, in terms of both number of operations and empirical runtimes.

Temporal Point Pattern Analysis of Human Activities Using GIS Methods: A Case Study of Library Visiting Activities in Chengdu City, China

This article aims to develop a new type of temporal pattern analysis—temporal point pattern analysis (TPPA)—by treating the distribution of activities as a point pattern on a two-dimensional plane using the start time and end time of activities as axes. Geographic information systems (GIS) methods, which are originally used in spatial point pattern analysis in GIS, are introduced to support TPPA. This article presents a case study to understand the temporal patterns of the library visiting activities of university students using a four-week smart card data set in Chengdu City, China. Several methods from GIS are applied, including the measurement of mean centers, kernel density, nearest neighbor distances, and optimized hot spot analysis. Results show that the GIS methods can reveal a lot of information on the temporal pattern of activities, thereby proving the reasonability of the proposed TPPA of activities. Key Words: GIS, human activities, library visiting, smart card data, visualization.

Video surveillance system against anti-terrorism by Using Adaptive Linear Activity Classification (ALAC) Technique.

Automated human activity analysis has been, and remains, a challenging problem. Security and surveillance are essential issues in today's world. Any behavior which is uncommon in occurrence and deviates from customarily understood action could be termed as suspicious. For different application regions, while identifying human exercises, fundamentally three angles are taking in worry for human movement recognition system: Segmentation, feature extraction, and activity classification. This model aims at automatic detection of abnormal behavior in surveillance videos. In this proposed work adaptive linear activity classification method and internet of things (IoT) frameworks are used to detection human activities as well as to find out who is doing unusual activities. The enhanced plan of the built environment condition will give a better observation. Such framework can be actualized in peoples in general places, for example, shopping centers, airports, and railway station or any private premises where security is the prime concern. The proposed ALAC method validated through simulation using MATLAB and VB.net software. Its ability to detect the activity of human the simulation result shows the effectiveness using ALAC method, Overall 97% efficiency achieved by using ALAC method.

A Literature Review: Geometric Methods and Their Applications in Human-Related Analysis.

Geometric features, such as the topological and manifold properties, are utilized to extract geometric properties. Geometric methods that exploit the applications of geometrics, e.g., geometric features, are widely used in computer graphics and computer vision problems. This review presents a literature review on geometric concepts, geometric methods, and their applications in human-related analysis, e.g., human shape analysis, human pose analysis, and human action analysis. This review proposes to categorize geometric methods based on the scope of the geometric properties that are extracted: object-oriented geometric methods, feature-oriented geometric methods, and routine-based geometric methods. Considering the broad applications of deep learning methods, this review also studies geometric deep learning, which has recently become a popular topic of research. Validation datasets are collected, and method performances are collected and compared. Finally, research trends and possible research topics are discussed.

RGB-D sensing based human action and interaction analysis: A survey

Human activity recognition has been actively studied in the last three decades. Compared to human action performed by a single person, human interaction is more complex due to the involvement of more subjects and the interdependence between them. Recently, motivated by the remarkable success of deep learning techniques, many learning-based feature representations have been developed for activity recognition. This paper provides a comprehensive review of human action and interaction recognition methods, covering both hand-crafted features and learning-based features, with a special focus on data captured by RGB-D sensors. Furthermore, this review reveals practical challenges in human activity analysis along with their promising solutions and potential future directions.

A Survey of Human Action Analysis in HRI Applications

The human action is an important information source for human social interaction, and it simultaneously plays a crucial role in human–robot interaction (HRI). For a natural and fluent interaction, robots are required to understand human actions and have the capacity to predict action intentions and to imitate human actions for an appropriate response. Currently, existing survey papers for the action recognition mainly summarize algorithms that perform action recognition in experimental scenarios, and survey papers of the HRI mainly introduced various interaction interfaces in the HRI. Different from these surveys, we focus on the human action analysis on robot platforms for the HRI application, including the body motion and gestures. We review the existing HRI related references involving the action recognition, prediction, and the robot imitation of the human action. Moreover, we give a summary of robot platforms and action datasets that are frequently used in the study of HRI. Finally, we give an analysis on the development trend and future research directions of action analysis for the HRI applications.

Face detection in the operating room: comparison of state-of-the-art methods and a self-supervised approach.

Face detection is a needed component for the automatic analysis and assistance of human activities during surgical procedures. Efficient face detection algorithms can indeed help to detect and identify the persons present in the room and also be used to automatically anonymize the data. However, current algorithms trained on natural images do not generalize well to the operating room (OR) images. In this work, we provide a comparison of state-of-the-art face detectors on OR data and also present an approach to train a face detector for the OR by exploiting non-annotated OR images. We propose a comparison of six state-of-the-art face detectors on clinical data using multi-view OR faces, a dataset of OR images capturing real surgical activities. We then propose to use self-supervision, a domain adaptation method, for the task of face detection in the OR. The approach makes use of non-annotated images to fine-tune a state-of-the-art detector for the OR without using any human supervision. The results show that the best model, namely the tiny face detector, yields an average precision of 0.556 at intersection over union of 0.5. Our self-supervised model using non-annotated clinical data outperforms this result by 9.2%. We present the first comparison of state-of-the-art face detectors on OR images and show that results can be significantly improved by using self-supervision on non-annotated data.

Policing as Digital Platform

Much of the discussion about platforms and “platform capitalism” centers on commercial platform companies like Google, Facebook, Amazon, and Apple. Shoshana Zuboff’s (2015) analysis of “surveillance capitalism” similarly focuses on Google as the trailblazer pushing the new logic of accumulation that is focused on data extraction and analysis of human activities. In his typology of platform companies, Nick Srnicek (2017) includes less visible industrial platforms that situate themselves as intermediaries between companies rather than between companies and consumer-users. In this article, the focus is a platform-building effort that looks something like an industrial platform but differs in the sense that the company in question, Axon Enterprise, aims to situate itself as an intermediary within and among law enforcement agencies (non-market entities) as a means of building a large-scale data-extractive system of monetization. Axon’s business strategy is emblematic of the ways that police evidence and record-keeping systems are being reimagined, and to some extent reconfigured, as sources of data extraction and analytics on the model of the platform. Whether Axon succeeds or is eclipsed by a competitor like Palantir or even Amazon or Microsoft, the process of reimagining and reorganizing policing as a platform is underway—a process that, to paraphrase Zuboff, deeply imbricates public and private surveillance activities, dissolving the boundary between public and private authority in the surveillance project.