@article{Mallol2022,

title = {Outer Product-Based Fusion of Smartwatch Sensor Data for Human Activity Recognition},

author = {Adria Mallol-Ragolta and Anastasia Semertzidou and Maria Pateraki and Bj\"{o}rn Schuller},

url = {https://www.frontiersin.org/articles/10.3389/fcomp.2022.796866},

doi = {10.3389/fcomp.2022.796866},

issn = {2624-9898},

year  = {2022},

date = {2022-01-01},

urldate = {2022-01-01},

journal = {Frontiers in Computer Science},

volume = {4},

abstract = {The advent of IoT devices in combination with Human Activity Recognition (HAR) technologies can contribute to battle with sedentariness by continuously monitoring the users' daily activities. With this information, autonomous systems could detect users' physical weaknesses and plan personalized training routines to improve them. This work investigates the multimodal fusion of smartwatch sensor data for HAR. Specifically, we exploit pedometer, heart rate, and accelerometer information to train unimodal and multimodal models for the task at hand. The models are trained end-to-end, and we compare the performance of dedicated Recurrent Neural Network-based (RNN) and Convolutional Neural Network-based (CNN) architectures to extract deep learnt representations from the input modalities. To fuse the embedded representations when training the multimodal models, we investigate a concatenation-based and an outer product-based approach. This work explores the harAGE dataset, a new dataset for HAR collected using a Garmin Vivoactive 3 device with more than 17 h of data. Our best models obtain an Unweighted Average Recall (UAR) of 95.6, 69.5, and 60.8% when tackling the task as a 2-class, 7-class, and 10-class classification problem, respectively. These performances are obtained using multimodal models that fuse the embedded representations extracted with dedicated CNN-based architectures from the pedometer, heart rate, and accelerometer modalities. The concatenation-based fusion scores the highest UAR in the 2-class classification problem, while the outer product-based fusion obtains the best performances in the 7-class and the 10-class classification problems.},

keywords = {artificial intelligence, Human Activity Recognition, multimodal fusion, smartwatch sensor data, ubiquitous computing},

pubstate = {published},

tppubtype = {article}

}

@article{MallolRagolta2022sustAGE1,

title = {sustAGE 1.0 \textendash First Prototype, Use Cases, and Usability Evaluation},

author = {Adria Mallol-Ragolta and Iraklis Varlamis and Maria Pateraki and Manolis I. A. Lourakis and Georgios Athanassiou and Michail Maniadakis and Konstantinos E. Papoutsakis and Thodoris Papadopoulos and Anastasia Semertzidou and Nicholas Cummins and Bj\"{o}rn Schuller and Ion-Anastasios Karolos and Christos Pikridas and Petros G. Patias and Spyros Vantolas and Leonidas Kallipolitis and Frank Werner and Antonio Ascolese and Vito Nitti},

url = {https://openaccess-api.cms-conferences.org/articles/download/978-1-7923-8989-4_58},

year  = {2022},

date = {2022-01-01},

urldate = {2022-01-01},

journal = {AHFE International},

abstract = {Worldwide demographics are changing; we are living longer and, in developed countries, the birth-rate is dropping. In this context and motivated by the challenge of sustainable ageing, this paper presents sustAGE, a multi-modal person-centred IoT platform, which integrates with the daily activities of ageing employees both at work and outside. The sensed information allows the system to assess the state of the users and context-related aspects with the aim to provide timely recommendations to support wellbeing, wellness, and productivity. Herein, we describe the use cases, outline the overall system architecture, and introduce the first prototype of the platform implemented up-to-date. Furthermore, the results from the usability evaluation conducted with real users who used the prototype for one month are presented},

keywords = {Ageing Workforce, artificial intelligence, IoT, Micro-moments, Occupational Safety and Health, Personalised Recommendations},

pubstate = {published},

tppubtype = {article}

}

@article{Antoniou2020,

title = {A deep learning method to accelerate the disaster response process},

author = {Vyron Antoniou and Chryssy Potsiou},

url = {https://www.unisdr.org/},

doi = {10.3390/rs12030544},

issn = {20724292},

year  = {2020},

date = {2020-02-01},

journal = {Remote Sens.},

volume = {12},

number = {3},

pages = {544},

publisher = {MDPI AG},

abstract = {This paper presents an end-to-end methodology that can be used in the disaster response process. The core element of the proposed method is a deep learning process which enables a helicopter landing site analysis through the identification of soccer fields. The method trains a deep learning autoencoder with the help of volunteered geographic information and satellite images. The process is mostly automated, it was developed to be applied in a time-and resource-constrained environment and keeps the human factor in the loop in order to control the final decisions. We show that through this process the cognitive load (CL) for an expert image analyst will be reduced by 70%, while the process will successfully identify 85.6% of the potential landing sites. We conclude that the suggested methodology can be used as part of a disaster response process.},

keywords = {artificial intelligence, deep learning, Deep learning autoencoder, Disaster response management, Helicopter landing site analysis, Machine Learning, Object detection, Satellite imagery, Volunteered geographic information},

pubstate = {published},

tppubtype = {article}

}