A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices
Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvi...
Ausführliche Beschreibung
Autor*in: |
Liu, Tianyi [verfasserIn] |
---|
Format: |
Artikel |
---|---|
Sprache: |
Englisch |
Erschienen: |
2021 |
---|
Schlagwörter: |
---|
Anmerkung: |
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 |
---|
Übergeordnetes Werk: |
Enthalten in: The journal of supercomputing - Springer US, 1987, 78(2021), 5 vom: 25. Okt., Seite 6696-6716 |
---|---|
Übergeordnetes Werk: |
volume:78 ; year:2021 ; number:5 ; day:25 ; month:10 ; pages:6696-6716 |
Links: |
---|
DOI / URN: |
10.1007/s11227-021-04140-5 |
---|
Katalog-ID: |
OLC2078292451 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | OLC2078292451 | ||
003 | DE-627 | ||
005 | 20230506001225.0 | ||
007 | tu | ||
008 | 221220s2021 xx ||||| 00| ||eng c | ||
024 | 7 | |a 10.1007/s11227-021-04140-5 |2 doi | |
035 | |a (DE-627)OLC2078292451 | ||
035 | |a (DE-He213)s11227-021-04140-5-p | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
082 | 0 | 4 | |a 004 |a 620 |q VZ |
100 | 1 | |a Liu, Tianyi |e verfasserin |4 aut | |
245 | 1 | 0 | |a A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
264 | 1 | |c 2021 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ohne Hilfsmittel zu benutzen |b n |2 rdamedia | ||
338 | |a Band |b nc |2 rdacarrier | ||
500 | |a © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 | ||
520 | |a Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. | ||
650 | 4 | |a Sensor | |
650 | 4 | |a Convolutional neural networks | |
650 | 4 | |a Activity recognition | |
650 | 4 | |a Deep learning | |
650 | 4 | |a Linear grouped convolution | |
700 | 1 | |a Wang, Shuoyuan |4 aut | |
700 | 1 | |a Liu, Yue |4 aut | |
700 | 1 | |a Quan, Weiming |4 aut | |
700 | 1 | |a Zhang, Lei |0 (orcid)0000-0001-8749-7459 |4 aut | |
773 | 0 | 8 | |i Enthalten in |t The journal of supercomputing |d Springer US, 1987 |g 78(2021), 5 vom: 25. Okt., Seite 6696-6716 |w (DE-627)13046466X |w (DE-600)740510-8 |w (DE-576)018667775 |x 0920-8542 |7 nnns |
773 | 1 | 8 | |g volume:78 |g year:2021 |g number:5 |g day:25 |g month:10 |g pages:6696-6716 |
856 | 4 | 1 | |u https://doi.org/10.1007/s11227-021-04140-5 |z lizenzpflichtig |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a SYSFLAG_A | ||
912 | |a GBV_OLC | ||
912 | |a SSG-OLC-TEC | ||
912 | |a SSG-OLC-MAT | ||
951 | |a AR | ||
952 | |d 78 |j 2021 |e 5 |b 25 |c 10 |h 6696-6716 |
author_variant |
t l tl s w sw y l yl w q wq l z lz |
---|---|
matchkey_str |
article:09208542:2021----::lgtegterlewrfaeoksnlnagopdovltofruaatvt |
hierarchy_sort_str |
2021 |
publishDate |
2021 |
allfields |
10.1007/s11227-021-04140-5 doi (DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p DE-627 ger DE-627 rakwb eng 004 620 VZ Liu, Tianyi verfasserin aut A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices 2021 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution Wang, Shuoyuan aut Liu, Yue aut Quan, Weiming aut Zhang, Lei (orcid)0000-0001-8749-7459 aut Enthalten in The journal of supercomputing Springer US, 1987 78(2021), 5 vom: 25. Okt., Seite 6696-6716 (DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 0920-8542 nnns volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 https://doi.org/10.1007/s11227-021-04140-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 78 2021 5 25 10 6696-6716 |
spelling |
10.1007/s11227-021-04140-5 doi (DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p DE-627 ger DE-627 rakwb eng 004 620 VZ Liu, Tianyi verfasserin aut A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices 2021 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution Wang, Shuoyuan aut Liu, Yue aut Quan, Weiming aut Zhang, Lei (orcid)0000-0001-8749-7459 aut Enthalten in The journal of supercomputing Springer US, 1987 78(2021), 5 vom: 25. Okt., Seite 6696-6716 (DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 0920-8542 nnns volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 https://doi.org/10.1007/s11227-021-04140-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 78 2021 5 25 10 6696-6716 |
allfields_unstemmed |
10.1007/s11227-021-04140-5 doi (DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p DE-627 ger DE-627 rakwb eng 004 620 VZ Liu, Tianyi verfasserin aut A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices 2021 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution Wang, Shuoyuan aut Liu, Yue aut Quan, Weiming aut Zhang, Lei (orcid)0000-0001-8749-7459 aut Enthalten in The journal of supercomputing Springer US, 1987 78(2021), 5 vom: 25. Okt., Seite 6696-6716 (DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 0920-8542 nnns volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 https://doi.org/10.1007/s11227-021-04140-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 78 2021 5 25 10 6696-6716 |
allfieldsGer |
10.1007/s11227-021-04140-5 doi (DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p DE-627 ger DE-627 rakwb eng 004 620 VZ Liu, Tianyi verfasserin aut A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices 2021 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution Wang, Shuoyuan aut Liu, Yue aut Quan, Weiming aut Zhang, Lei (orcid)0000-0001-8749-7459 aut Enthalten in The journal of supercomputing Springer US, 1987 78(2021), 5 vom: 25. Okt., Seite 6696-6716 (DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 0920-8542 nnns volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 https://doi.org/10.1007/s11227-021-04140-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 78 2021 5 25 10 6696-6716 |
allfieldsSound |
10.1007/s11227-021-04140-5 doi (DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p DE-627 ger DE-627 rakwb eng 004 620 VZ Liu, Tianyi verfasserin aut A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices 2021 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution Wang, Shuoyuan aut Liu, Yue aut Quan, Weiming aut Zhang, Lei (orcid)0000-0001-8749-7459 aut Enthalten in The journal of supercomputing Springer US, 1987 78(2021), 5 vom: 25. Okt., Seite 6696-6716 (DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 0920-8542 nnns volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 https://doi.org/10.1007/s11227-021-04140-5 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 78 2021 5 25 10 6696-6716 |
language |
English |
source |
Enthalten in The journal of supercomputing 78(2021), 5 vom: 25. Okt., Seite 6696-6716 volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 |
sourceStr |
Enthalten in The journal of supercomputing 78(2021), 5 vom: 25. Okt., Seite 6696-6716 volume:78 year:2021 number:5 day:25 month:10 pages:6696-6716 |
format_phy_str_mv |
Article |
institution |
findex.gbv.de |
topic_facet |
Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution |
dewey-raw |
004 |
isfreeaccess_bool |
false |
container_title |
The journal of supercomputing |
authorswithroles_txt_mv |
Liu, Tianyi @@aut@@ Wang, Shuoyuan @@aut@@ Liu, Yue @@aut@@ Quan, Weiming @@aut@@ Zhang, Lei @@aut@@ |
publishDateDaySort_date |
2021-10-25T00:00:00Z |
hierarchy_top_id |
13046466X |
dewey-sort |
14 |
id |
OLC2078292451 |
language_de |
englisch |
fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2078292451</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230506001225.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">221220s2021 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11227-021-04140-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2078292451</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s11227-021-04140-5-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="a">620</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Liu, Tianyi</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2021</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Sensor</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Convolutional neural networks</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Activity recognition</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Deep learning</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Linear grouped convolution</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Wang, Shuoyuan</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Liu, Yue</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Quan, Weiming</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Zhang, Lei</subfield><subfield code="0">(orcid)0000-0001-8749-7459</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">The journal of supercomputing</subfield><subfield code="d">Springer US, 1987</subfield><subfield code="g">78(2021), 5 vom: 25. Okt., Seite 6696-6716</subfield><subfield code="w">(DE-627)13046466X</subfield><subfield code="w">(DE-600)740510-8</subfield><subfield code="w">(DE-576)018667775</subfield><subfield code="x">0920-8542</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:78</subfield><subfield code="g">year:2021</subfield><subfield code="g">number:5</subfield><subfield code="g">day:25</subfield><subfield code="g">month:10</subfield><subfield code="g">pages:6696-6716</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s11227-021-04140-5</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-TEC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">78</subfield><subfield code="j">2021</subfield><subfield code="e">5</subfield><subfield code="b">25</subfield><subfield code="c">10</subfield><subfield code="h">6696-6716</subfield></datafield></record></collection>
|
author |
Liu, Tianyi |
spellingShingle |
Liu, Tianyi ddc 004 misc Sensor misc Convolutional neural networks misc Activity recognition misc Deep learning misc Linear grouped convolution A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
authorStr |
Liu, Tianyi |
ppnlink_with_tag_str_mv |
@@773@@(DE-627)13046466X |
format |
Article |
dewey-ones |
004 - Data processing & computer science 620 - Engineering & allied operations |
delete_txt_mv |
keep |
author_role |
aut aut aut aut aut |
collection |
OLC |
remote_str |
false |
illustrated |
Not Illustrated |
issn |
0920-8542 |
topic_title |
004 620 VZ A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices Sensor Convolutional neural networks Activity recognition Deep learning Linear grouped convolution |
topic |
ddc 004 misc Sensor misc Convolutional neural networks misc Activity recognition misc Deep learning misc Linear grouped convolution |
topic_unstemmed |
ddc 004 misc Sensor misc Convolutional neural networks misc Activity recognition misc Deep learning misc Linear grouped convolution |
topic_browse |
ddc 004 misc Sensor misc Convolutional neural networks misc Activity recognition misc Deep learning misc Linear grouped convolution |
format_facet |
Aufsätze Gedruckte Aufsätze |
format_main_str_mv |
Text Zeitschrift/Artikel |
carriertype_str_mv |
nc |
hierarchy_parent_title |
The journal of supercomputing |
hierarchy_parent_id |
13046466X |
dewey-tens |
000 - Computer science, knowledge & systems 620 - Engineering |
hierarchy_top_title |
The journal of supercomputing |
isfreeaccess_txt |
false |
familylinks_str_mv |
(DE-627)13046466X (DE-600)740510-8 (DE-576)018667775 |
title |
A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
ctrlnum |
(DE-627)OLC2078292451 (DE-He213)s11227-021-04140-5-p |
title_full |
A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
author_sort |
Liu, Tianyi |
journal |
The journal of supercomputing |
journalStr |
The journal of supercomputing |
lang_code |
eng |
isOA_bool |
false |
dewey-hundreds |
000 - Computer science, information & general works 600 - Technology |
recordtype |
marc |
publishDateSort |
2021 |
contenttype_str_mv |
txt |
container_start_page |
6696 |
author_browse |
Liu, Tianyi Wang, Shuoyuan Liu, Yue Quan, Weiming Zhang, Lei |
container_volume |
78 |
class |
004 620 VZ |
format_se |
Aufsätze |
author-letter |
Liu, Tianyi |
doi_str_mv |
10.1007/s11227-021-04140-5 |
normlink |
(ORCID)0000-0001-8749-7459 |
normlink_prefix_str_mv |
(orcid)0000-0001-8749-7459 |
dewey-full |
004 620 |
title_sort |
a lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
title_auth |
A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
abstract |
Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 |
abstractGer |
Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 |
abstract_unstemmed |
Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021 |
collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT |
container_issue |
5 |
title_short |
A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices |
url |
https://doi.org/10.1007/s11227-021-04140-5 |
remote_bool |
false |
author2 |
Wang, Shuoyuan Liu, Yue Quan, Weiming Zhang, Lei |
author2Str |
Wang, Shuoyuan Liu, Yue Quan, Weiming Zhang, Lei |
ppnlink |
13046466X |
mediatype_str_mv |
n |
isOA_txt |
false |
hochschulschrift_bool |
false |
doi_str |
10.1007/s11227-021-04140-5 |
up_date |
2024-07-03T19:44:25.667Z |
_version_ |
1803588327878688768 |
fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2078292451</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230506001225.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">221220s2021 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11227-021-04140-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2078292451</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s11227-021-04140-5-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="a">620</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Liu, Tianyi</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">A lightweight neural network framework using linear grouped convolution for human activity recognition on mobile devices</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2021</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Human activity recognition (HAR) has played an indispensable role in ubiquitous computing scenario, from smart homes to game console designing, elderly care, and fitness tracking. It is very hard to manually extract most suitable activity features from raw sensor time series. Due to an obvious advantage, convolutional neural networks (CNNs) that can extract features automatically have been widely utilized for activity recognition. Despite exceptional performance, CNNs are computation-intensive and memory-demanding algorithms because of a large number of internal parameters. As a result, the research attention in HAR implementations over resource-limited embedded platforms has turned to computationally lightweight convolution architectures. In this paper, we offer a contribution in the direction. Simple linear transformations with low cost are combined with a convolution-based HAR classifier to decrease overall computational/memory overhead and, simultaneously, which establishes an efficient classification method with satisfactory performance. Our new method is evaluated against standard convolution-based and residual counterparts, over several popular HAR datasets for algorithm benchmarking. Experimental results verify that, the cheap linear operations can significantly reduce computational and memory cost, and meanwhile producing satisfactory recognition performance, which is able to ensure faster inference on mobile devices. Our new method could be a strong candidate for real HAR implementations on embedded platforms with limited computing resources.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Sensor</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Convolutional neural networks</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Activity recognition</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Deep learning</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Linear grouped convolution</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Wang, Shuoyuan</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Liu, Yue</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Quan, Weiming</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Zhang, Lei</subfield><subfield code="0">(orcid)0000-0001-8749-7459</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">The journal of supercomputing</subfield><subfield code="d">Springer US, 1987</subfield><subfield code="g">78(2021), 5 vom: 25. Okt., Seite 6696-6716</subfield><subfield code="w">(DE-627)13046466X</subfield><subfield code="w">(DE-600)740510-8</subfield><subfield code="w">(DE-576)018667775</subfield><subfield code="x">0920-8542</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:78</subfield><subfield code="g">year:2021</subfield><subfield code="g">number:5</subfield><subfield code="g">day:25</subfield><subfield code="g">month:10</subfield><subfield code="g">pages:6696-6716</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s11227-021-04140-5</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-TEC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">78</subfield><subfield code="j">2021</subfield><subfield code="e">5</subfield><subfield code="b">25</subfield><subfield code="c">10</subfield><subfield code="h">6696-6716</subfield></datafield></record></collection>
|
score |
7.399811 |