OmniIMU-Bench: Benchmarking Wrist IMU Representations Across Kinematic, Semantic and Clinical Tasks

Anonymous (ICLR 2027 double-blind submission)

Summary

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.