Karan Ahuja

Papers

• OmniIMU-Bench: Benchmarking Wrist IMU Representations Across Kinematic, Semantic and Clinical Tasks
  2026 · ICLR 2027 submission

  Benchmark of ten tasks from eight datasets for wrist IMU multi-task performance and cross-task transfer (activity, gesture, pose & velocity estimation, odometry, clinical motor assessment, IMU–language retrieval). Compares six SSL methods pretrained on unlabelled recorded and simulated wrist motion. Masked autoencoding is best/joint-best frozen-probe on 9/10 tasks; fine-tuning beats scratch on 9/10.

  • Venue: ICLR 2027 submission (under review); public release as IMU-Bench (Li, Bradshaw, Gao, Ahuja 2026; arXiv 2610.19481, submitted 2026-10-14)
  • Signals: ACC, GYRO
  • Form factor: wrist
• SensorFM: Toward a General Intelligence and Interface for Wearable Health Data
  2026 · arXiv preprint

  Google's SensorFM is a wearable health foundation model pretrained on more than one trillion minutes of unlabelled sensor data from five million participants. Scaling model size and data together steadily improved results across 35 health prediction tasks in cardiovascular, metabolic, sleep and mental health, lifestyle and demographics.

  • Published: arXiv preprint, 2026
  • Pretraining: 1 trillion+ minutes of sensor data
  • Cohort: 5 million participants

Models

• InrtlWrist: masked-autoencoder wrist IMU encoder (IMU-Bench reference model)
• SensorFM (Google)

Datasets used in this work

• OpenWatch: smartwatch IMU+PPG hand-gesture benchmark (n=50, >10 h, 59 gesture sequences)
• HHAR: heterogeneity HAR, smartphones + smartwatches (n=9)
• mhealth
• WISDM AR v1.1: cellphone accelerometer HAR (n=36)
• PAMAP2: Physical Activity Monitoring (n=9, 3× IMU + HR)

Co-authors