Bio-Inspired Self-Supervised Learning for Wrist-worn IMU Signals

• URL: http://arxiv.org/abs/2603.10961v1
• Date: Wed, 11 Mar 2026 16:48:33 GMT
• Title: Bio-Inspired Self-Supervised Learning for Wrist-worn IMU Signals
• Abstract summary: We introduce a novel tokenization strategy grounded in the submovement theory of motor control.<n>We define our token as the movement segment, a unit of motion composed of a finite sequence of submovements.<n>Our representations outperform strong wearable SSL baselines across six subject-disjoint HAR benchmarks.
• Score: 21.295169769759053
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Wearable accelerometers have enabled large-scale health and wellness monitoring, yet learning robust human-activity representations has been constrained by the scarcity of labeled data. While self-supervised learning offers a potential remedy, existing approaches treat sensor streams as unstructured time series, overlooking the underlying biological structure of human movement, a factor we argue is critical for effective Human Activity Recognition (HAR). We introduce a novel tokenization strategy grounded in the submovement theory of motor control, which posits that continuous wrist motion is composed of superposed elementary basis functions called submovements. We define our token as the movement segment, a unit of motion composed of a finite sequence of submovements that is readily extractable from wrist accelerometer signals. By treating these segments as tokens, we pretrain a Transformer encoder via masked movement-segment reconstruction to model the temporal dependencies of movement segments, shifting the learning focus beyond local waveform morphology. Pretrained on the NHANES corpus (approximately 28k hours; approximately 11k participants; approximately 10M windows), our representations outperform strong wearable SSL baselines across six subject-disjoint HAR benchmarks. Furthermore, they demonstrate stronger data efficiency in data-scarce settings. Code and pretrained weights will be made publicly available.

Related papers

• Classifying daily activities needs posture, reconstructing them needs motion [0.0]
  We used videos of 16 daily activities from the MoVi dataset to compare three movement analysis strategies.<n>We found two discriminative features for movement classification.<n>We identified 9 critical joints that are most predictive for movement classification.
  arXiv  Detail & Related papers  (2026-07-14T19:21:32Z)
• MotionMAR: Multi-scale Auto-Regressive Human Motion Reconstruction from Sparse Observations [51.837372131319285]
  MotionMAR is a framework for motion reconstruction from sparse observations.<n>It first estimates the global trajectory of human motion and then gradually refines the temporal details.<n>MotionMAR achieves state-of-the-art accuracy on the AMASS dataset.
  arXiv  Detail & Related papers  (2026-06-22T08:15:56Z)
• MotionTeller: Multi-modal Integration of Wearable Time-Series with LLMs for Health and Behavioral Understanding [4.158479111055355]
  MotionTeller is a generative framework that integrates minute-level wearable activity data with large language models (LLMs)<n>We construct a novel dataset of 54383 (actigraphy, text) pairs derived from real-world NHANES recordings, and train the model using cross-entropy loss with supervision only on the language tokens.<n>MotionTeller achieves high semantic fidelity (BERT-F1 = 0.924) and lexical accuracy (ROUGE-1 = 0.722), outperforming prompt-based baselines by 7 percent in ROUGE-1.
  arXiv  Detail & Related papers  (2025-12-25T04:37:07Z)
• Deep Learning for Motion Classification in Ankle Exoskeletons Using Surface EMG and IMU Signals [0.8388591755871735]
  Ankle exoskeletons have garnered considerable interest for their potential to enhance mobility and reduce fall risks. This paper presents a novel motion prediction framework that integrates three Inertial Measurement Units (IMUs) and eight surface Electromyography (sEMG) sensors. Our findings reveal that Convolutional Neural Networks (CNNs) slightly outperform Long Short-Term Memory (LSTM) networks on a dataset of five motion tasks.
  arXiv  Detail & Related papers  (2024-11-25T10:51:40Z)
• Scaling Wearable Foundation Models [54.93979158708164]
  We investigate the scaling properties of sensor foundation models across compute, data, and model size. Using a dataset of up to 40 million hours of in-situ heart rate, heart rate variability, electrodermal activity, accelerometer, skin temperature, and altimeter per-minute data from over 165,000 people, we create LSM. Our results establish the scaling laws of LSM for tasks such as imputation, extrapolation, both across time and sensor modalities.
  arXiv  Detail & Related papers  (2024-10-17T15:08:21Z)
• Masked Motion Predictors are Strong 3D Action Representation Learners [143.9677635274393]
  In 3D human action recognition, limited supervised data makes it challenging to fully tap into the modeling potential of powerful networks such as transformers. We show that instead of following the prevalent pretext to perform masked self-component reconstruction in human joints, explicit contextual motion modeling is key to the success of learning effective feature representation for 3D action recognition.
  arXiv  Detail & Related papers  (2023-08-14T11:56:39Z)
• Spatio-Temporal Branching for Motion Prediction using Motion Increments [55.68088298632865]
  Human motion prediction (HMP) has emerged as a popular research topic due to its diverse applications. Traditional methods rely on hand-crafted features and machine learning techniques. We propose a noveltemporal-temporal branching network using incremental information for HMP.
  arXiv  Detail & Related papers  (2023-08-02T12:04:28Z)
• Human Activity Recognition on wrist-worn accelerometers using self-supervised neural networks [0.0]
  Measures of Activity of Daily Living (ADL) are an important indicator of overall health but difficult to measure in-clinic. We propose a self-supervised learning paradigm to create a robust representation of accelerometer data that can generalize across devices and subjects. We also propose a segmentation algorithm which can identify segments of salient activity and boost HAR accuracy on continuous real-life data.
  arXiv  Detail & Related papers  (2021-12-22T23:35:20Z)
• Continuous Decoding of Daily-Life Hand Movements from Forearm Muscle Activity for Enhanced Myoelectric Control of Hand Prostheses [78.120734120667]
  We introduce a novel method, based on a long short-term memory (LSTM) network, to continuously map forearm EMG activity onto hand kinematics. Ours is the first reported work on the prediction of hand kinematics that uses this challenging dataset. Our results suggest that the presented method is suitable for the generation of control signals for the independent and proportional actuation of the multiple DOFs of state-of-the-art hand prostheses.
  arXiv  Detail & Related papers  (2021-04-29T00:11:32Z)
• Segmentation and Classification of EMG Time-Series During Reach-to-Grasp Motion [10.388787606334745]
  We propose a framework for classifying EMG signals generated from continuous grasp movements with variations on dynamic arm/hand postures. The proposed framework was evaluated in real-time with the accuracy variation over time presented.
  arXiv  Detail & Related papers  (2021-04-19T20:41:06Z)
• Learning Generalizable Physiological Representations from Large-scale Wearable Data [12.863826659440026]
  We present a novel self-supervised representation learning method using activity and heart rate (HR) signals without semantic labels. We show that the resulting embeddings can generalize in various downstream tasks through transfer learning with linear classifiers. Overall, we propose the first multimodal self-supervised method for behavioral and physiological data with implications for large-scale health and lifestyle monitoring.
  arXiv  Detail & Related papers  (2020-11-09T17:56:03Z)
• Learning Motion Flows for Semi-supervised Instrument Segmentation from Robotic Surgical Video [64.44583693846751]
  We study the semi-supervised instrument segmentation from robotic surgical videos with sparse annotations. By exploiting generated data pairs, our framework can recover and even enhance temporal consistency of training sequences. Results show that our method outperforms the state-of-the-art semisupervised methods by a large margin.
  arXiv  Detail & Related papers  (2020-07-06T02:39:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.