WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition
IntroductionPhysics simulation has emerged as a promising approach to generate virtual Inertial Measurement Unit (IMU) data, offering a solution to reduce the extensive cost and effort of real-world data collection. However, the fidelity of virtual IMU depends heavily on the quality of the source mo...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Computer Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcomp.2025.1514933/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590867582418944 |
|---|---|
| author | Nobuyuki Oishi Phil Birch Daniel Roggen Paula Lago |
| author_facet | Nobuyuki Oishi Phil Birch Daniel Roggen Paula Lago |
| author_sort | Nobuyuki Oishi |
| collection | DOAJ |
| description | IntroductionPhysics simulation has emerged as a promising approach to generate virtual Inertial Measurement Unit (IMU) data, offering a solution to reduce the extensive cost and effort of real-world data collection. However, the fidelity of virtual IMU depends heavily on the quality of the source motion data, which varies with motion capture setups. We hypothesize that improving virtual IMU fidelity is crucial to fully harness the potential of physics simulation for virtual IMU data generation in training Human Activity Recognition (HAR) models.MethodTo investigate this, we introduce WIMUSim, a 6-axis wearable IMU simulation framework designed to accurately parameterize real IMU properties when deployed on people. WIMUSim models IMUs in wearable sensing using four key parameters: Body (skeletal model), Dynamics (movement patterns), Placement (device positioning), and Hardware (IMU characteristics). Using these parameters, WIMUSim simulates virtual IMU through differentiable vector manipulations and quaternion rotations. A key novelty enabled by this approach is the identification of WIMUSim parameters using recorded real IMU data through gradient descent-based optimization, starting from an initial estimate. This process enhances the fidelity of the virtual IMU by optimizing the parameters to closely mimic the recorded IMU data. Adjusting these identified parameters allows us to introduce physically plausible variabilities.ResultsOur fidelity assessment demonstrates that WIMUSim accurately replicates real IMU data with optimized parameters and realistically simulates changes in sensor placement. Evaluations using exercise and locomotion activity datasets confirm that models trained with optimized virtual IMU data perform comparably to those trained with real IMU data. Moreover, we demonstrate the use of WIMUSim for data augmentation through two approaches: Comprehensive Parameter Mixing, which enhances data diversity by varying parameter combinations across subjects, outperforming models trained with real and non-optimized virtual IMU data by 4–10 percentage points (pp); and Personalized Dataset Generation, which customizes augmented datasets to individual user profiles, resulting in average accuracy improvements of 4 pp, with gains exceeding 10 pp for certain subjects.DiscussionThese results underscore the benefit of high-fidelity virtual IMU data and WIMUSim's utility in developing effective data generation strategies, alleviating the challenge of data scarcity in sensor-based HAR. |
| format | Article |
| id | doaj-art-1f07bbe7fa904d0086324f097729e72d |
| institution | Kabale University |
| issn | 2624-9898 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computer Science |
| spelling | doaj-art-1f07bbe7fa904d0086324f097729e72d2025-01-23T06:56:06ZengFrontiers Media S.A.Frontiers in Computer Science2624-98982025-01-01710.3389/fcomp.2025.15149331514933WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognitionNobuyuki Oishi0Phil Birch1Daniel Roggen2Paula Lago3School of Engineering and Informatics, University of Sussex, Brighton, United KingdomSchool of Engineering and Informatics, University of Sussex, Brighton, United KingdomSchool of Engineering and Informatics, University of Sussex, Brighton, United KingdomDepartment of Electrical and Computer Engineering, Concordia University, Montreal, QC, CanadaIntroductionPhysics simulation has emerged as a promising approach to generate virtual Inertial Measurement Unit (IMU) data, offering a solution to reduce the extensive cost and effort of real-world data collection. However, the fidelity of virtual IMU depends heavily on the quality of the source motion data, which varies with motion capture setups. We hypothesize that improving virtual IMU fidelity is crucial to fully harness the potential of physics simulation for virtual IMU data generation in training Human Activity Recognition (HAR) models.MethodTo investigate this, we introduce WIMUSim, a 6-axis wearable IMU simulation framework designed to accurately parameterize real IMU properties when deployed on people. WIMUSim models IMUs in wearable sensing using four key parameters: Body (skeletal model), Dynamics (movement patterns), Placement (device positioning), and Hardware (IMU characteristics). Using these parameters, WIMUSim simulates virtual IMU through differentiable vector manipulations and quaternion rotations. A key novelty enabled by this approach is the identification of WIMUSim parameters using recorded real IMU data through gradient descent-based optimization, starting from an initial estimate. This process enhances the fidelity of the virtual IMU by optimizing the parameters to closely mimic the recorded IMU data. Adjusting these identified parameters allows us to introduce physically plausible variabilities.ResultsOur fidelity assessment demonstrates that WIMUSim accurately replicates real IMU data with optimized parameters and realistically simulates changes in sensor placement. Evaluations using exercise and locomotion activity datasets confirm that models trained with optimized virtual IMU data perform comparably to those trained with real IMU data. Moreover, we demonstrate the use of WIMUSim for data augmentation through two approaches: Comprehensive Parameter Mixing, which enhances data diversity by varying parameter combinations across subjects, outperforming models trained with real and non-optimized virtual IMU data by 4–10 percentage points (pp); and Personalized Dataset Generation, which customizes augmented datasets to individual user profiles, resulting in average accuracy improvements of 4 pp, with gains exceeding 10 pp for certain subjects.DiscussionThese results underscore the benefit of high-fidelity virtual IMU data and WIMUSim's utility in developing effective data generation strategies, alleviating the challenge of data scarcity in sensor-based HAR.https://www.frontiersin.org/articles/10.3389/fcomp.2025.1514933/fullphysics simulationoptimizationinertial measurement unitwearable computinghuman activity recognition |
| spellingShingle | Nobuyuki Oishi Phil Birch Daniel Roggen Paula Lago WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition Frontiers in Computer Science physics simulation optimization inertial measurement unit wearable computing human activity recognition |
| title | WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition |
| title_full | WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition |
| title_fullStr | WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition |
| title_full_unstemmed | WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition |
| title_short | WIMUSim: simulating realistic variabilities in wearable IMUs for human activity recognition |
| title_sort | wimusim simulating realistic variabilities in wearable imus for human activity recognition |
| topic | physics simulation optimization inertial measurement unit wearable computing human activity recognition |
| url | https://www.frontiersin.org/articles/10.3389/fcomp.2025.1514933/full |
| work_keys_str_mv | AT nobuyukioishi wimusimsimulatingrealisticvariabilitiesinwearableimusforhumanactivityrecognition AT philbirch wimusimsimulatingrealisticvariabilitiesinwearableimusforhumanactivityrecognition AT danielroggen wimusimsimulatingrealisticvariabilitiesinwearableimusforhumanactivityrecognition AT paulalago wimusimsimulatingrealisticvariabilitiesinwearableimusforhumanactivityrecognition |