Exploring Artificial Intelligence methods for recognizing human activities in real time by exploiting inertial sensors
| Autor: | Boucharas, Dimitrios, Androutsos, Christos, Tachos, Nikolaos S., Tripoliti, Evanthia E., Manousos, Dimitrios, Skaramagkas, Vasileios, Ktistakis, Emmanouil, Tsiknakis, Manolis, Fotiadis, Dimitrios I. |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | 2021 IEEE 21st International Conference on Bioinformatics and Bioengineering (BIBE) |
| DOI: | 10.5281/zenodo.5704960 |
| Popis: | The aim of this work is to present two different algorithmic pipelines for human activity recognition (HAR) in real time, exploiting inertial measurement unit (IMU) sensors. Various learning classifiers have been developed and tested across different datasets. The experimental results provide a comparative performance analysis based on accuracy and latency during fine-tuning, training and prediction. The overall accuracy of the proposed pipeline reaches 66% in the publicly available dataset and 90% in the in-house one. Pre-final version (preprint). For the citation of this journal paper please use the DOI provided of the publisher (IEEE) which is the following one DOI:https://doi.org/10.1109/BIBE52308.2021.9635486 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|