Fu, Biying ; Damer, Naser ; Kirchbuchner, Florian ; Kuijper, Arjan (2021)
Generalization of Fitness Exercise Recognition from Doppler Measurements by Domain-Adaption and Few-Shot Learning.
25th Interantional Conference on Pattern Recognition (ICPR 2021). virtual Conference (10.-11.01.2021)
doi: 10.1007/978-3-030-68799-1_15
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
In previous works, a mobile application was developed using an unmodified commercial smartphone to recognize whole-body exercises. The working principle was based on the ultrasound Doppler sensing with the device built-in hardware. Applying such a lab environment trained model on realistic application variations causes a significant drop in performance, and thus decimate its applicability. The reason of the reduced performance can be manifold. It could be induced by the user, environment, and device variations in realistic scenarios. Such scenarios are often more complex and diverse, which can be challenging to anticipate in the initial training data. To study and overcome this issue, this paper presents a database with controlled and uncontrolled subsets of fitness exercises. We propose two concepts to utilize small adaption data to successfully improve model generalization in an uncontrolled environment, increasing the recognition accuracy by two to six folds compared to the baseline for different users.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2021 |
| Autor(en): | Fu, Biying ; Damer, Naser ; Kirchbuchner, Florian ; Kuijper, Arjan |
| Art des Eintrags: | Bibliographie |
| Titel: | Generalization of Fitness Exercise Recognition from Doppler Measurements by Domain-Adaption and Few-Shot Learning |
| Sprache: | Englisch |
| Publikationsjahr: | 2021 |
| Verlag: | Springer |
| Buchtitel: | Pattern Recognition : ICPR International Workshops and Challenges |
| Veranstaltungstitel: | 25th Interantional Conference on Pattern Recognition (ICPR 2021) |
| Veranstaltungsort: | virtual Conference |
| Veranstaltungsdatum: | 10.-11.01.2021 |
| DOI: | 10.1007/978-3-030-68799-1_15 |
| Kurzbeschreibung (Abstract): | In previous works, a mobile application was developed using an unmodified commercial smartphone to recognize whole-body exercises. The working principle was based on the ultrasound Doppler sensing with the device built-in hardware. Applying such a lab environment trained model on realistic application variations causes a significant drop in performance, and thus decimate its applicability. The reason of the reduced performance can be manifold. It could be induced by the user, environment, and device variations in realistic scenarios. Such scenarios are often more complex and diverse, which can be challenging to anticipate in the initial training data. To study and overcome this issue, this paper presents a database with controlled and uncontrolled subsets of fitness exercises. We propose two concepts to utilize small adaption data to successfully improve model generalization in an uncontrolled environment, increasing the recognition accuracy by two to six folds compared to the baseline for different users. |
| Freie Schlagworte: | Human activity recognition, Mobile sensors, Physical activity monitoring, Domain adaptation |
| Zusätzliche Informationen: | Proceedings Part I; Part of the Lecture Notes in Computer Science book series (LNCS, volume 12661) |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme 20 Fachbereich Informatik > Mathematisches und angewandtes Visual Computing |
| Hinterlegungsdatum: | 16 Mär 2021 08:23 |
| Letzte Änderung: | 16 Mär 2021 08:23 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum