Keypoints-Based Multi-Cue Feature Fusion Network (MF-Net) for Action Recognition of ADHD Children in TOVA Assessment
Wanyu Tang, Chao Shi, Yuanyuan Li, Zhonglan Tang, Gang Yang, Jing Zhang, Ling He · 2024
Read Keypoints-Based Multi-Cue Feature Fusion Network (MF-Net) for Action Recognition of ADHD Children in TOVA Assessment by Wanyu Tang, Chao Shi, Yuanyuan Li, Zhonglan Tang, Gang Yang, Jing Zhang, Ling He (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11672943.txt.
Description
Wanyu Tang and co-authors present a 2024 system that reads the actions of children with ADHD from video. The children take a standard attention test while a camera watches. Sudden, uneven moves are a hallmark of the condition, and set video tools often miss them. The team's system tracks body and face points instead of raw pixels, which cuts background noise.
One branch reads body points with a graph network tuned for jerky motion. A second branch turns face points into images and reads head and face moves with a phone-grade vision model. A fusion step joins both branches for the final call on each act. The calls map to the symptom traits in the standard diagnostic guide. Such tools could make attention tests more fair and less reliant on a watcher's eye.
Catalog record
| CREATOR | Wanyu Tang, Chao Shi, Yuanyuan Li, Zhonglan Tang, Gang Yang, Jing Zhang, Ling He |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |