PaperarXiv:2611.19027
HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition
IEEE BigData 2026 (accepted; arXiv extended version 2611.19027, submitted 2026-11-13)
TL;DR
Sensor-aware SSL framework pretrained on eight wearable datasets (173 subjects, 358,261 windows) that keeps temporal modelling within each sensor separate from cross-sensor modelling and simply omits missing sensors. Under leave-one-dataset-out pretraining and subject-disjoint tests, mean macro-F1 rises from 0.688 to 0.724 over the same architecture trained from scratch; with half the channels removed at test time it keeps 73.8% of full performance vs 64.1%. MIT code, no released weights.
Abstract
Self-supervised learning for wearable sensing is usually confined to a single dataset or device family, leaving open whether representations transfer across heterogeneous sensor configurations and whether gains come from pretraining rather than architecture or richer sensor input. We present HeteroWear, a sensor-aware self-supervised framework pretrained on an eight-dataset wearable corpus (173 subjects, 358,261 windows) spanning inertial and physiological signals across body locations, devices, and sampling rates. HeteroWear separates within-sensor temporal modeling from cross-sensor relational modeling using sensor-context embeddings, temporal and cross-sensor Transformers, and latent cross-attention fusion. Training combines structured sensor dropout, cross-configuration consistency, and per-axis masked sensor modeling; missing sensors are omitted rather than reconstructed. Under leave-one-dataset-out pretraining and subject-disjoint evaluation on PAMAP2, WISDM, HHAR, HARTH, and UCI-HAR, HeteroWear improves mean macro-F1 from 0.688 to 0.724 over the identical architecture trained from scratch, with the largest gain on HHAR (0.823 versus 0.741), and achieves the highest macro-F1 on four of five benchmarks relative to single-accelerometer supervised baselines. Held-out-device evaluation yields a mean +0.066 macro-F1 improvement. Under 50% test-time channel removal, HeteroWear retains 73.8% of full-channel performance versus 64.1% for the same architecture trained from scratch without sensor dropout. Ablations show that structured sensor dropout primarily drives robustness, while cross-configuration consistency and sensor-context encoding contribute to transfer.
Verbatim from arXiv (opens in a new tab) · arXiv metadata (abstracts are released under CC0)
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Tasks
Signals
Devices
Keywords
Datasets used
PAMAP2Physical Activity Monitoring (n=9, 3× IMU + HR)PretrainingEvaluation9 subjects · IMU · Temp · HR
WISDM AR v1.1cellphone accelerometer HAR (n=36)PretrainingEvaluation36 subjects · IMU
HHARheterogeneity HAR, smartphones + smartwatches (n=9)PretrainingEvaluation9 subjects · IMU
HARTHthigh + lower-back accelerometers, free-living, video-annotated (n=22)PretrainingEvaluation22 subjects · IMU
UCI Human Activity Recognition Using Smartphones (UCI-HAR)PretrainingEvaluation30 subjects · IMU
PPG-DaLiAPPG dataset for motion compensation / HRPretraining15 subjects · PPG · ECG · IMU
WESADWearable Stress and Affect DetectionPretraining15 subjects · PPG · ECG · IMU
Models
How it works
Cite this paper
Mohammadpour, Y., & Shaeri, P. (2026). HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition. IEEE BigData 2026. https://arxiv.org/abs/2611.19027
Mohammadpour, Yasaman and Pouya Shaeri. 2026. "HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition." IEEE BigData 2026. https://arxiv.org/abs/2611.19027.
Mohammadpour, Y. and Shaeri, P. (2026) 'HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition', IEEE BigData 2026. Available at: https://arxiv.org/abs/2611.19027
Mohammadpour, Yasaman, and Pouya Shaeri "HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition." IEEE BigData 2026, 2026. https://arxiv.org/abs/2611.19027.
Y. Mohammadpour and P. Shaeri, "HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition," IEEE BigData 2026, 2026, [Online]. Available: https://arxiv.org/abs/2611.19027.
@article{mohammadpour2026heterowear,
title = {{HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition}},
author = {Mohammadpour, Yasaman and Shaeri, Pouya},
year = {2026},
journal = {IEEE BigData 2026},
eprint = {2611.19027},
archivePrefix = {arXiv}
}Added to the ChatPPG research hub