# Representing and Detecting Label Ambiguity in IMU-Based Exercise Evaluation

arXiv: https://arxiv.org/abs/2607.04842 · HTML: https://mlsanity.com/paper/2607.04842/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 30 references that are in the map.
- First posted on arXiv 6 Jul 2026.

## Builds on (most cited first, 5 of 5)

- [Deep Label Distribution Learning with Label Ambiguity](https://mlsanity.com/paper/1611.01731/) (2016)
- [Mitigating Label Noise through Data Ambiguation](https://mlsanity.com/paper/2305.13764/) (2023)
- [Automatic Assessment of Functional Movement Screening Exercises with Deep Learning Architectures](https://mlsanity.com/paper/2210.01209/) (2022)
- [Boosting Automatic Exercise Evaluation Through Musculoskeletal Simulation-Based IMU Data Augmentation](https://mlsanity.com/paper/2505.24415/) (2025)
- [GAITEX: Human motion dataset of impaired gait and rehabilitation exercises using inertial and optical sensors](https://mlsanity.com/paper/2507.21069/) (2025)

## Abstract (arXiv)

Home-based physiotherapy is performed without supervision, which leads to incorrect execution and motivates systems that assess movement automatically from inertial measurement units (IMUs). Such systems assign each repetition to a category, yet a relevant share of repetitions falls near a class boundary, where even trained raters disagree. Classifiers trained with one-hot labels collapse these borderline repetitions onto a single class and discard this ambiguity. We address this with a method that automatically generates a label distribution per repetition without a large rater pool. We train a network to reproduce the full distribution with a Kullback-Leibler objective, the ambiguity approach, and compare it against a one-hot cross-entropy baseline on four IMU exercise datasets. From the network output we further determine whether a repetition is ambiguous and which classes are relevant to it. The ambiguity approach matched or exceeded the baseline classification on all four datasets, and detected ambiguity and the relevant classes more reliably. Representing the label distribution in the training target therefore adds information about ambiguity at no cost to classification.