What is it about?
Wearable devices often struggle to recognize specific activities because most recorded data actually shows no clear activity at all. This work studies how to reduce the negative impact of this dominant “no activity” class when recognizing sports movements from wearable sensor data.
Featured Image
Photo by Solen Feyissa on Unsplash
Why is it important?
Better handling of “no activity” data makes wearable systems more accurate and trustworthy for real-world use in sports tracking, health monitoring, and fitness applications.
Read the Original
This page is a summary of: Mitigating Null-Class Dominance in Multiclass Inertial-Based Activity Recognition: HASCA-WEAR Challenge - A Technical Report of Team HARMA, October 2025, ACM (Association for Computing Machinery),
DOI: 10.1145/3714394.3756192.
You can read the full text:
Contributors
The following have contributed to this page
