Evaluation of Smartphone Camera Positioning on Artificial Intelligence Pose Estimation Accuracy for Exercise Detection: Observational Study
JMIR mHealth and uHealth · 05.03.2026 · 14 · e82412-e82412
DOI: 10.2196/82412 ↗Kurz gefasst
Beobachtungsstudie in der Zeitschrift JMIR mHealth and uHealth, online erschienen am 05.03.2026. Verfasst von Eduarda Oliosi und 7 weiteren Autor:innen. Beteiligte Einrichtung: Value for Health CoLAB, 15 Fontes Pereira de Melo Ave, 2nd Fl, Right, Lisbon, 1050‑115, Portugal, 351 937091767. Hauptthemen laut MEDLINE: Smartphone, Artificial Intelligence, Exercise. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Artificial intelligence (AI)–driven pose estimation (PE) offers a scalable and cost-effective solution to track exercises in mobile health apps. However, occlusion, influenced by camera angle and distance, can reduce detection accuracy and repetition counting precision. The influence of smartphone positioning on these performance metrics remains underexplored in controlled studies.
Objective
The study aimed to examine how smartphone camera angle (front, side, and diagonal) and distance (90 cm, 180 cm, 200 cm, and 360 cm) affect detection performance and repetition counting accuracy during push-ups and squats using AI-based PE.
Methods
In this cross-sectional, within-subject study, 44 healthy university students (9 [20.5%] female participants; mean age 20.3 y, SD 0.4 y; mean BMI 23.2, SD 0.6 kg/m²) were assigned to perform either squats or push-ups. Each participant completed their assigned exercise across 12 predefined smartphone camera configurations, yielding approximately 264 squat trials (n=22) and 264 push-up trials (n=22). Each trial consisted of an average of 5 repetitions, totaling approximately 1320 repetitions per exercise. PE performance was assessed using binary classification accuracy, detection rate, and mean absolute error (MAE) for repetition counting. Generalized linear mixed-effects models evaluated classification odds, linear mixed-effects models analyzed MAE, and Tukey-adjusted post hoc tests followed significant effects.
Results
The mean detection rate was 61.1% (SD 48.8%) for push-ups and 61.5% (SD 48.7%) for squats, with MAEs of 1.08 (SD 1.78) and 1.11 (SD 1.82) repetitions, respectively. Push-ups were most accurately detected from diagonal views at 90 to 180 cm (up to 85.7% detection; MAE=0.28) and least accurately from the front at 360 cm (20%; MAE=2.70). Squats performed best from a diagonal view at 200 cm (95.5%; MAE=0.05) and worst from the side at 90 cm (0%; MAE=5). Generalized linear mixed models showed that for push-ups, the front 90 cm and diagonal 360 cm views significantly reduced classification odds compared to the side 90 cm view ( P =.03 and P =.04, respectively), whereas for squats, diagonal and front views significantly outperformed side views across all distances ( P
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Value for Health CoLAB, 15 Fontes Pereira de Melo Ave, 2nd Fl, Right, Lisbon, 1050‑115, Portugal, 351 937091767.
- Laboratory for Instrumentation, Biomedical Engineering and Radiation Physics (LIBPhys‑UNL), Physics Department, Nova School of Science and Technology, Nova University of Lisbon, Caparica, Portugal.
- Research Centre for Physical Activity, Health and Leisure (CIAFEL), Faculty of Sport, University of Porto, Porto, Portugal.
- Departamento de Desporto e Saúde, Escola de Saúde e Desenvolvimento Humano, Universidade de Évora, Évora, Portugal.
- Comprehensive Health Research Centre (CHRC), Escola de Saúde e Desenvolvimento Humano, Departamento de Ciências Médicas, Universidade de Évora, Évora, Portugal.
- Research Unit in Business and Economics, Católica Lisbon School of Business and Economics (CUBE), Catholic University of Portugal, Lisbon, Portugal.
ORCID-Profile: Eduarda Oliosi, Soraia Ferreira, Ana Paula Giordano, Guilherme Viveiros, José Parraca, Paulo Pereira, Federico Guede-Fernández, Salomé Azevedo
Zitieren
APA
Oliosi, E., Ferreira, S., Giordano, A. P., Viveiros, G., Parraca, J., Pereira, P., Guede-Fernández, F., & Azevedo, S. (2026). Evaluation of Smartphone Camera Positioning on Artificial Intelligence Pose Estimation Accuracy for Exercise Detection: Observational Study. JMIR mHealth and uHealth, 14, e82412-e82412. https://doi.org/10.2196/82412
Vancouver
Oliosi E, Ferreira S, Giordano AP, Viveiros G, Parraca J, Pereira P, et al. Evaluation of Smartphone Camera Positioning on Artificial Intelligence Pose Estimation Accuracy for Exercise Detection: Observational Study. JMIR mHealth and uHealth. 2026;14:e82412-e82412. doi:10.2196/82412
BibTeX
@article{Oliosi2026,
author = {Oliosi, Eduarda and Ferreira, Soraia and Giordano, Ana Paula and Viveiros, Guilherme and Parraca, José and Pereira, Paulo and Guede-Fernández, Federico and Azevedo, Salomé},
title = {Evaluation of Smartphone Camera Positioning on Artificial Intelligence Pose Estimation Accuracy for Exercise Detection: Observational Study},
journal = {JMIR mHealth and uHealth},
year = {2026},
volume = {14},
pages = {e82412--e82412},
doi = {10.2196/82412}
}
EUVIMED