Building Lightweight Deep learning Models with TensorFlow Lite for Human Activity Recognition on Mobile Devices

Sevda Özge Bursa*, Özlem Durmaz İncel, Gülfem Işıklar Alptekin

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

2 Citations (Scopus)

Abstract

Human activity recognition (HAR) is a research domain that enables continuous monitoring of human behaviors for various purposes, from assisted living to surveillance in smart home environments. These applications generally work with a rich collection of sensor data generated using smartphones and other low-power wearable devices. The amount of collected data can quickly become immense, necessitating time and resource-consuming computations. Deep learning (DL) has recently become a promising trend in HAR. However, it is challenging to train and run DL algorithms on mobile devices due to their limited battery power, memory, and computation units. In this paper, we evaluate and compare the performance of four different deep architectures trained on three datasets from the HAR literature (WISDM, MobiAct, OpenHAR). We use the TensorFlow Lite platform with quantization techniques to convert the models into lighter versions for deployment on mobile devices. We compare the performance of the original models in terms of accuracy, size, and resource usage with their optimized versions. The experiments reveal that the model size and resource consumption can significantly be reduced when optimized with TensorFlow Lite without sacrificing the accuracy of the models.

Original languageEnglish
Pages (from-to)687-702
Number of pages16
JournalAnnales des Telecommunications/Annals of Telecommunications
Volume78
Issue number11-12
DOIs
Publication statusPublished - Nov 2023
Externally publishedYes

Keywords

  • n/a OA procedure
  • Energy consumption
  • Human activity recognition (HAR)
  • Resource-constrained devices
  • Wearable sensors
  • Deep learning (DL)

Fingerprint

Dive into the research topics of 'Building Lightweight Deep learning Models with TensorFlow Lite for Human Activity Recognition on Mobile Devices'. Together they form a unique fingerprint.

Cite this