A Multi-Agent Digital Twin Framework for AI-Driven Fitness Coaching
Open accessAbstract
Figure 1: The architecture of the proposed Digital Twin AI Fitness Coaching System (DTAIFC).The system integrates multimodal input sensing, short and long-term memory, a Crew-inspired multi-agent AI coordination core, and a user-facing interaction interface.Orchestration Agent operates as the central brain coordinating the Recommendation and Feedback Agents for intelligent workout planning and real-time adjustment.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Agentic AI is characterized by multi-step, multi-layered tasks over extended time frames.
“Multi-step, multi-layered tasks over extended time frames”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Fitness and wellness agents have coordinated recommendations.
“coordinated recommendations [60]”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Fitness and wellness agents have offered multimodal feedback.
“offered multimodal feedback”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Fitness and wellness agents have strengthened posture analysis paradigms.
“strengthened posture analysis paradigms”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Fitness and wellness agents have tracked progress.
“tracked progress”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Fitness and wellness agents have enabled adaptive coaching.
“Fitness and wellness agents have enabled adaptive coaching”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Feedback-generating interactive systems can use audio to goal parse and make context-aware outputs.
“Similarly, feedback-generating interactive systems can use audio to goal parse and make context-aware outputs [60].”
From The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges (2025) · cites this paper as [60]
Cites
Advances in Neural Information Processing Systems 19
(2007) · The MIT Press eBooks · doi:10.52202/085713 · Cited by 16,037 · see citing works
OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields
Cao, Z., Hidalgo, G., Simon, T., Wei, S.-E., & Sheikh, Y. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2929257 · Cited by 5,536 · see citing works
Digital Twins: The Convergence of Multimedia Technologies
El Saddik, A. (2018) · IEEE MultiMedia · doi:10.1109/mmul.2018.023121167 · Cited by 817 · see citing works
Digital Twin Coaching for Physical Activities: A Survey
Díaz, R. G., Yu, Q., Ding, Y., Laamarti, F., & Saddik, A. E. (2020) · Italian National Conference on Sensors · doi:10.3390/s20205936 · Cited by 81 · see citing works
Self-citationKowatsch, T., Lohse, K.-M., Erb, V., Schittenhelm, L., Galliker, H., Lehner, R., & Huang, E. M. (2021) · Journal of Medical Internet Research · doi:10.2196/23612 · Cited by 39 · see citing works
Spitale, M., Axelsson, M., & Güneş, H. (2025) · ACM Transactions on Human-Robot Interaction · doi:10.1145/3712265 · Cited by 30 · see citing works
eXtended meta-uni-omni-Verse (XV): Introduction, Taxonomy, and State-of-the-Art
Mann, S., Yuan, Y., Lamberti, F., Saddik, A. E., Thawonmas, R., & Pratticò, F. G. (2023) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2023.3283728 · Cited by 29 · see citing works
The Integration of ChatGPT With the Metaverse for Medical Consultations
Saddik, A. E., & Ghaboura, S. (2023) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2023.3324978 · Cited by 27 · see citing works
Zhang, J., Han, D., Han, S., Li, H., Lam, W., & Zhang, M. (2024) · Computer Speech & Language · doi:10.1016/j.csl.2024.101694 · Cited by 25 · see citing works
Meta-Review on Brain-Computer Interface (BCI) in the Metaverse
HamlAbadi, K. G., Laamarti, F., & Saddik, A. E. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3696109 · Cited by 23 · see citing works
Self-citationMeta-Review of Wearable Devices for Healthcare in the Metaverse
Vahdati, M., Laamarti, F., & Saddik, A. E. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3705320 · Cited by 11 · see citing works
Self-citationMultimodal Agents: From Vision to Reality
Jain, R. (2024) · IEEE Multimedia · doi:10.1109/mmul.2024.3485253 · Cited by 7 · see citing works
Digital Twin Coaching for Edge Computing Using Deep Learning Based 2D Pose Estimation
Díaz, R. G. (2021) · uO Research (University of Ottawa) · doi:10.20381/ruor-26229 · Cited by 3 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Bandi, A., Kongari, B., Naguru, R., Pasnoor, S., & Vilipala, S. V. (2025) · Future Internet · doi:10.3390/fi17090404 · Cited by 207 · see citing works
HamlAbadi, K. G., Vahdati, M., Laamarti, F., & Saddik, A. E. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3758176 · Cited by 3 · see citing works
Self-citationDigital Twin AI Fitness Coach: An Intelligent Multi-Agent System for Personalized Exercise Guidance
Vahdati, M., HamlAbadi, K. G., Laamarti, F., Kumar, D., & Saddik, A. E. (2025) · doi:10.1145/3728485.3759171 · Cited by 1 · see citing works
Self-citationFrom Talking Heads to Moving Bodies: Understanding Video-Mediated Dance Instruction
Eriksson, S., McMillan, D., & Brown, B. (2026) · ACM International Conference on Interactive Media Experiences · doi:10.1145/3788851.3805021
An Agentic AI-Based Prototype to Support Teaching-Learning Processes
Flores-Serrano, D., & Chicaiza, J. A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-32112-1_12
- Authors
- Vahdati, M. (., Gholizadeh HamlAbadi, K., Laamarti, F., & El Saddik, A.
- Affiliations
- Multimedia Communications Research Laboratory (MCRLab), School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, Ontario, Canada
- University of Ottawa (CA)
- Venue
- Proceedings of the 2025 ACM International Conference on Interactive Media Experiences
- Publisher
- ACM
- Published
- 2025-05-31
- Pages
- 380-385
- OpenAlex
- W4410919314
- Cited by
- 5
- References
- 13
- Citations by year
- 2026: 2 · 2025: 3
- FWCI
- 3.78
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM International Conference on Interactive Media Experiences (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Health Professions › Health Information Management › Innovation in Digital Healthcare Systems
- Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
- Keywords
- digital twin, multi-agent systems, exercise coaching, workout planning, multimodal sensing, memory architecture, agent coordination, recommendation agents
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
