LARa: Creating a Dataset for Human Activity Recognition in Logistics Using Semantic Attributes

Promoting fairness in activity recognition algorithms for patient's monitoring and evaluation systems in healthcare

Researchers face the challenge of defining subject selection criteria when training algorithms for human activity recognition tasks. The ongoing uncertainty revolves around which characteristics should be considered to ensure algorithmic robustness across diverse populations. This study aims to address this challenge by conducting an analysis of heterogeneity in the training data to assess the impact of physical characteristics and soft-biometric attributes on activity recognition performance. The performance of various state-of-the-art deep neural network architectures (tCNN, hybrid-LSTM, Transformer model) processing time-series data using the IntelliRehab (IRDS) dataset was evaluated. By intentionally introducing bias into the training data based on human characteristics, the objective is to identify the characteristics that influence algorithms in motion analysis. Experimental findings reveal that the CNN-LSTM model achieved the highest accuracy, reaching 88%. Moreover, models trained on heterogeneous distributions of disability attributes exhibited notably higher accuracy, reaching 51%, compared to those not considering such factors, which scored an average of 33%. These evaluations underscore the significant influence of subjects' characteristics on activity recognition performance, providing valuable insights into the algorithm's robustness across diverse populations. This study represents a significant step forward in promoting fairness and trustworthiness in artificial intelligence by quantifying representation bias in multi-channel time-series activity recognition data within the healthcare domain.

Biosensor-Based Multimodal Deep Human Locomotion Decoding via Internet of Healthcare Things

Multiple Internet of Healthcare Things (IoHT)-based devices have been utilized as sensing methodologies for human locomotion decoding to aid in applications related to e-healthcare. Different measurement conditions affect the daily routine monitoring, including the sensor type, wearing style, data retrieval method, and processing model. Currently, several models are present in this domain that include a variety of techniques for pre-processing, descriptor extraction, and reduction, along with the classification of data captured from multiple sensors. However, such models consisting of multiple subject-based data using different techniques may degrade the accuracy rate of locomotion decoding. Therefore, this study proposes a deep neural network model that not only applies the state-of-the-art Quaternion-based filtration technique for motion and ambient data along with background subtraction and skeleton modeling for video-based data, but also learns important descriptors from novel graph-based representations and Gaussian Markov random-field mechanisms. Due to the non-linear nature of data, these descriptors are further utilized to extract the codebook via the Gaussian mixture regression model. Furthermore, the codebook is provided to the recurrent neural network to classify the activities for the locomotion-decoding system. We show the validity of the proposed model across two publicly available data sampling strategies, namely, the HWU-USP and LARa datasets. The proposed model is significantly improved over previous systems, as it achieved 82.22% and 82.50% for the HWU-USP and LARa datasets, respectively. The proposed IoHT-based locomotion-decoding model is useful for unobtrusive human activity recognition over extended periods in e-healthcare facilities.

Less is more: Efficient behavioral context recognition using Dissimilarity-Based Query Strategy

With the advancement of ubiquitous computing, smartphone sensors are generating a vast amount of unlabeled data streams ubiquitously. This sensor data can potentially help to recognize various behavioral contexts in the natural environment. Accurate behavioral context recognition has a wide variety of applications in many domains like disease prevention and independent living. However, despite the availability of enormous amounts of sensor data, label acquisition, due to its dependence on users, is still a challenging task. In this work, we propose a novel context recognition approach i.e., Dissimilarity-Based Query Strategy (DBQS). Our approach DBQS leverages Active Learning based selective sampling to find the informative and diverse samples in the sensor data to train the model. Our approach overcomes the stagnation problem by considering only new and distinct samples from the pool that were not previously explored. Further, our model exploits temporal information in the data in order to further maintain diversity in the dataset. The key intuition behind the proposed approach is that the variations during the learning phase will train the model in diverse settings and it will outperform when assigned a context recognition task in the natural setting. Experimentation on a publicly available natural environment dataset demonstrates that our proposed approach improved overall average Balanced Accuracy(BA) by 6% with an overall 13% less training data requirement.

Review on Human Action Recognition in Smart Living: Sensing Technology, Multimodality, Real-Time Processing, Interoperability, and Resource-Constrained Processing

Smart living, a concept that has gained increasing attention in recent years, revolves around integrating advanced technologies in homes and cities to enhance the quality of life for citizens. Sensing and human action recognition are crucial aspects of this concept. Smart living applications span various domains, such as energy consumption, healthcare, transportation, and education, which greatly benefit from effective human action recognition. This field, originating from computer vision, seeks to recognize human actions and activities using not only visual data but also many other sensor modalities. This paper comprehensively reviews the literature on human action recognition in smart living environments, synthesizing the main contributions, challenges, and future research directions. This review selects five key domains, i.e., Sensing Technology, Multimodality, Real-time Processing, Interoperability, and Resource-Constrained Processing, as they encompass the critical aspects required for successfully deploying human action recognition in smart living. These domains highlight the essential role that sensing and human action recognition play in successfully developing and implementing smart living solutions. This paper serves as a valuable resource for researchers and practitioners seeking to further explore and advance the field of human action recognition in smart living.

Computer-assisted approaches for measuring, segmenting, and analyzing functional upper extremity movement: a narrative review of the current state, limitations, and future directions

The analysis of functional upper extremity (UE) movement kinematics has implications across domains such as rehabilitation and evaluating job-related skills. Using movement kinematics to quantify movement quality and skill is a promising area of research but is currently not being used widely due to issues associated with cost and the need for further methodological validation. Recent developments by computationally-oriented research communities have resulted in potentially useful methods for evaluating UE function that may make kinematic analyses easier to perform, generally more accessible, and provide more objective information about movement quality, the importance of which has been highlighted during the COVID-19 pandemic. This narrative review provides an interdisciplinary perspective on the current state of computer-assisted methods for analyzing UE kinematics with a specific focus on how to make kinematic analyses more accessible to domain experts. We find that a variety of methods exist to more easily measure and segment functional UE movement, with a subset of those methods being validated for specific applications. Future directions include developing more robust methods for measurement and segmentation, validating these methods in conjunction with proposed kinematic outcome measures, and studying how to integrate kinematic analyses into domain expert workflows in a way that improves outcomes.

Ergonomic human-robot collaboration in industry: A review

In the current industrial context, the importance of assessing and improving workers' health conditions is widely recognised. Both physical and psycho-social factors contribute to jeopardising the underlying comfort and well-being, boosting the occurrence of diseases and injuries, and affecting their quality of life. Human-robot interaction and collaboration frameworks stand out among the possible solutions to prevent and mitigate workplace risk factors. The increasingly advanced control strategies and planning schemes featured by collaborative robots have the potential to foster fruitful and efficient coordination during the execution of hybrid tasks, by meeting their human counterparts' needs and limits. To this end, a thorough and comprehensive evaluation of an individual's ergonomics, i.e. direct effect of workload on the human psycho-physical state, must be taken into account. In this review article, we provide an overview of the existing ergonomics assessment tools as well as the available monitoring technologies to drive and adapt a collaborative robot's behaviour. Preliminary attempts of ergonomic human-robot collaboration frameworks are presented next, discussing state-of-the-art limitations and challenges. Future trends and promising themes are finally highlighted, aiming to promote safety, health, and equality in worldwide workplaces.

Recognizing Solo Jazz Dance Moves Using a Single Leg-Attached Inertial Wearable Device

We present here a method for recognising dance moves in sequences using 3D accelerometer and gyroscope signals, acquired by a single wearable device, attached to the dancer's leg. The recognition entails dance tempo estimation, temporal scaling, a wearable device orientation-invariant coordinate system transformation, and, finally, sliding correlation-based template matching. The recognition is independent of the orientation of the wearable device and the tempo of dancing, which promotes the usability of the method in a wide range of everyday application scenarios. For experimental validation, we considered the versatile repertoire of solo jazz dance moves. We created a database of 15 authentic solo jazz template moves using the performances of a professional dancer dancing at 120 bpm. We analysed 36 new dance sequences, performed by the professional and five recreational dancers, following six dance tempos, ranging from 120 bpm to 220 bpm with 20 bpm increment steps. The recognition F1 scores, obtained cumulatively for all moves for different tempos, ranged from 0.87 to 0.98. The results indicate that the presented method can be used to recognise repeated dance moves and to assess the dancer's consistency in performance. In addition, the results confirm the potential of using the presented method to recognise imitated dance moves, supporting the learning process.

Context-Aware Human Activity Recognition in Industrial Processes

The automatic, sensor-based assessment of human activities is highly relevant for production and logistics, to optimise the economics and ergonomics of these processes. One challenge for accurate activity recognition in these domains is the context-dependence of activities: Similar movements can correspond to different activities, depending on, e.g., the object handled or the location of the subject. In this paper, we propose to explicitly make use of such context information in an activity recognition model. Our first contribution is a publicly available, semantically annotated motion capturing dataset of subjects performing order picking and packaging activities, where context information is recorded explicitly. The second contribution is an activity recognition model that integrates movement data and context information. We empirically show that by using context information, activity recognition performance increases substantially. Additionally, we analyse which of the pieces of context information is most relevant for activity recognition. The insights provided by this paper can help others to design appropriate sensor set-ups in real warehouses for time management.

Inertial Measurement Unit Sensors in Assistive Technologies for Visually Impaired People, a Review

A diverse array of assistive technologies have been developed to help Visually Impaired People (VIP) face many basic daily autonomy challenges. Inertial measurement unit sensors, on the other hand, have been used for navigation, guidance, and localization but especially for full body motion tracking due to their low cost and miniaturization, which have allowed the estimation of kinematic parameters and biomechanical analysis for different field of applications. The aim of this work was to present a comprehensive approach of assistive technologies for VIP that include inertial sensors as input, producing results on the comprehension of technical characteristics of the inertial sensors, the methodologies applied, and their specific role in each developed system. The results show that there are just a few inertial sensor-based systems. However, these sensors provide essential information when combined with optical sensors and radio signals for navigation and special application fields. The discussion includes new avenues of research, missing elements, and usability analysis, since a limitation evidenced in the selected articles is the lack of user-centered designs. Finally, regarding application fields, it has been highlighted that a gap exists in the literature regarding aids for rehabilitation and biomechanical analysis of VIP. Most of the findings are focused on navigation and obstacle detection, and this should be considered for future applications.