mlsanityData to 14 Oct 2026

RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable Data

arXiv 2411.18822 ↗First posted 27 Nov 2024eess.SP cs.AI cs.LGcited by 20

Not on the map. Why?

Abstract (arXiv)

We present RelCon, a novel self-supervised Relative Contrastive learning approach for training a motion foundation model from wearable accelerometry sensors. First, a learnable distance measure is trained to capture motif similarity and domain-specific semantic information such as rotation invariance. Then, the learned distance provides a measurement of semantic similarity between a pair of accelerometry time-series, which we use to train our foundation model to model relative relationships across time and across subjects. The foundation model is trained on 1 billion segments from 87,376 participants, and achieves state-of-the-art performance across multiple downstream tasks, including human activity recognition and gait metric regression. To our knowledge, we are the first to show the generalizability of a foundation model with motion data from wearables across distinct evaluation tasks.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Builds on 39 of its 89 references that are in the map. First posted on arXiv 27 Nov 2024.

Builds on

39 of 89 references.

References · 39

  1. 2017Attention Is All You Needcited by 60,361
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  6. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  7. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  8. 2023A decoder-only foundation model for time-series forecastingcited by 692
  9. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 482
  10. 2021TS2Vec: Towards Universal Representation of Time Seriescited by 471
  11. 2016From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classificationcited by 341
  12. 2022Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistencycited by 250
27 more
  1. 2021Unsupervised Representation Learning for Time Series with Temporal Neighborhood Codingcited by 224
  2. 2020CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patientscited by 171
  3. 2020Boosting Contrastive Self-Supervised Learning with False Negative Cancellationcited by 119
  4. 2019Multi-task Self-Supervised Learning for Human Activity Detectioncited by 102
  5. 2023Large-scale Training of Foundation Models for Wearable Biosignalscited by 95
  6. 2020Exploring Contrastive Learning in Human Activity Recognition for Healthcarecited by 75
  7. 2022Self-supervised Learning for Human Activity Recognition Using 700,000 Person-days of Wearable Datacited by 75
  8. 2024PaPaGei: Open Foundation Models for Optical Physiological Signalscited by 69
  9. 2022Assessing the State of Self-Supervised Human Activity Recognition using Wearablescited by 65
  10. 2024Scaling Wearable Foundation Modelscited by 63
  11. 2020Contrastive Predictive Coding for Human Activity Recognitioncited by 55
  12. 2023Soft Contrastive Learning for Time Seriescited by 51
  13. 2018Protecting Sensory Data against Sensitive Inferencescited by 45
  14. 2024A Large-Scale Evaluation of Speech Foundation Modelscited by 40
  15. 2019Deep Metric Learning Beyond Binary Supervisioncited by 36
  16. 20213KG: Contrastive Learning of 12-Lead Electrocardiograms using Physiologically-Inspired Augmentationscited by 36
  17. 2021Patient Contrastive Learning: a Performant, Expressive, and Practical Approach to ECG Modelingcited by 25
  18. 2023REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learningcited by 21
  19. 2022Motiflets -- Simple and Accurate Detection of Motifs in Time Seriescited by 16
  20. 2023Event-Based Contrastive Learning for Medical Time Seriescited by 13
  21. 2022TASKED: Transformer-based Adversarial learning for human activity recognition using wearable sensors via Self-KnowledgE Distillationcited by 10
  22. 2023Towards Learning Discrete Representations via Self-Supervision for Wearables-Based Human Activity Recognitioncited by 7
  23. 2023Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signalscited by 4
  24. 2024Inertial Sensors for Human Motion Analysis: A Comprehensive Reviewcited by 4
  25. 2024Nightbeat: Heart Rate Estimation From a Wrist-Worn Accelerometer During Sleepcited by 4
  26. 2023Time-Series Contrastive Learning against False Negatives and Class Imbalancecited by 3
  27. 2018DeepWalking: Enabling Smartphone-based Walking Speed Estimation Using Deep Learningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 13.

20 papers.

Sort:

By area:

  1. 2025Pulse-PPG: An Open-Source Field-Trained PPG Foundation Model for Wearable Applications Across Lab and Field Settingscited by 37
  2. 2025LSM-2: Learning from Incomplete Wearable Sensor Datacited by 23
  3. 2025Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictionscited by 18
  4. 2025Foundation Models for CPS-IoT: Opportunities and Challengescited by 5
  5. 2026Reinforcement Learning in the Real World: A Survey of Statistical Challenges and Future Directionscited by 5
  6. 2026A robust PPG foundation model using multimodal physiological supervisioncited by 3
  7. 2026Inertia-1: An Open Exploration of Wearable Motion Foundation Modelscited by 3
  8. 2025SlotFM: A Motion Foundation Model with Slot Attention for Diverse Downstream Taskscited by 2
  9. 2026Bio-Inspired Self-Supervised Learning for Wrist-worn Accelerometer Datacited by 2
  10. 2026Sonata: A Hybrid World Model for Inertial Kinematics under Clinical Data Scarcitycited by 2
  11. 2026BenchHAR: Benchmarking Self-Supervised Learning for Generalizable Sensor-based Activity Recognitioncited by 2
  12. 2026AnyMo: Geometry-Aware Setup-Agnostic Modeling of Human Motion in the Wildcited by 1
8 more
  1. 2025Speech Foundation Models Generalize to Time Series Tasks from Wearable Sensor Datacited by 0
  2. 2025Prototype-based Self-Supervised Multimodal Learning for PPG and Accelerometry Signalscited by 0
  3. 2026Motion Capture is Not the Target Domain: Scaling Synthetic Data for Learning Motion Representationscited by 0
  4. 2026WavesFM: Hierarchical Representation Learning for Longitudinal Wearable Sensor Waveformscited by 0
  5. 2026VCR: Learning Valid Contextual Representation for Incomplete Wearable Signalscited by 0
  6. 2026Variable Input Masked Autoencoders for Wrist-Worn Accelerometrycited by 0
  7. 2026IMU-Bench: Benchmarking Wrist IMU Representations Across Kinematic, Semantic and Clinical Taskscited by 0
  8. 2026TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Seriescited by 0