Cargando…

On-Device Deep Learning Inference for Efficient Activity Data Collection

Labeling activity data is a central part of the design and evaluation of human activity recognition systems. The performance of the systems greatly depends on the quantity and “quality” of annotations; therefore, it is inevitable to rely on users and to keep them motivated to provide activity labels...

Descripción completa

Detalles Bibliográficos
Autores principales: Mairittha, Nattaya, Mairittha, Tittaya, Inoue, Sozo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6696120/
https://www.ncbi.nlm.nih.gov/pubmed/31387314
http://dx.doi.org/10.3390/s19153434
_version_ 1783444195736813568
author Mairittha, Nattaya
Mairittha, Tittaya
Inoue, Sozo
author_facet Mairittha, Nattaya
Mairittha, Tittaya
Inoue, Sozo
author_sort Mairittha, Nattaya
collection PubMed
description Labeling activity data is a central part of the design and evaluation of human activity recognition systems. The performance of the systems greatly depends on the quantity and “quality” of annotations; therefore, it is inevitable to rely on users and to keep them motivated to provide activity labels. While mobile and embedded devices are increasingly using deep learning models to infer user context, we propose to exploit on-device deep learning inference using a long short-term memory (LSTM)-based method to alleviate the labeling effort and ground truth data collection in activity recognition systems using smartphone sensors. The novel idea behind this is that estimated activities are used as feedback for motivating users to collect accurate activity labels. To enable us to perform evaluations, we conduct the experiments with two conditional methods. We compare the proposed method showing estimated activities using on-device deep learning inference with the traditional method showing sentences without estimated activities through smartphone notifications. By evaluating with the dataset gathered, the results show our proposed method has improvements in both data quality (i.e., the performance of a classification model) and data quantity (i.e., the number of data collected) that reflect our method could improve activity data collection, which can enhance human activity recognition systems. We discuss the results, limitations, challenges, and implications for on-device deep learning inference that support activity data collection. Also, we publish the preliminary dataset collected to the research community for activity recognition.
format Online
Article
Text
id pubmed-6696120
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-66961202019-09-05 On-Device Deep Learning Inference for Efficient Activity Data Collection Mairittha, Nattaya Mairittha, Tittaya Inoue, Sozo Sensors (Basel) Article Labeling activity data is a central part of the design and evaluation of human activity recognition systems. The performance of the systems greatly depends on the quantity and “quality” of annotations; therefore, it is inevitable to rely on users and to keep them motivated to provide activity labels. While mobile and embedded devices are increasingly using deep learning models to infer user context, we propose to exploit on-device deep learning inference using a long short-term memory (LSTM)-based method to alleviate the labeling effort and ground truth data collection in activity recognition systems using smartphone sensors. The novel idea behind this is that estimated activities are used as feedback for motivating users to collect accurate activity labels. To enable us to perform evaluations, we conduct the experiments with two conditional methods. We compare the proposed method showing estimated activities using on-device deep learning inference with the traditional method showing sentences without estimated activities through smartphone notifications. By evaluating with the dataset gathered, the results show our proposed method has improvements in both data quality (i.e., the performance of a classification model) and data quantity (i.e., the number of data collected) that reflect our method could improve activity data collection, which can enhance human activity recognition systems. We discuss the results, limitations, challenges, and implications for on-device deep learning inference that support activity data collection. Also, we publish the preliminary dataset collected to the research community for activity recognition. MDPI 2019-08-05 /pmc/articles/PMC6696120/ /pubmed/31387314 http://dx.doi.org/10.3390/s19153434 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Mairittha, Nattaya
Mairittha, Tittaya
Inoue, Sozo
On-Device Deep Learning Inference for Efficient Activity Data Collection
title On-Device Deep Learning Inference for Efficient Activity Data Collection
title_full On-Device Deep Learning Inference for Efficient Activity Data Collection
title_fullStr On-Device Deep Learning Inference for Efficient Activity Data Collection
title_full_unstemmed On-Device Deep Learning Inference for Efficient Activity Data Collection
title_short On-Device Deep Learning Inference for Efficient Activity Data Collection
title_sort on-device deep learning inference for efficient activity data collection
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6696120/
https://www.ncbi.nlm.nih.gov/pubmed/31387314
http://dx.doi.org/10.3390/s19153434
work_keys_str_mv AT mairitthanattaya ondevicedeeplearninginferenceforefficientactivitydatacollection
AT mairitthatittaya ondevicedeeplearninginferenceforefficientactivitydatacollection
AT inouesozo ondevicedeeplearninginferenceforefficientactivitydatacollection