Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention
Fan Yang, Shiyu Li, Chang Sun, Xingjiang Li, Zhangbo Xiao · 2024
Read Action recognition in rehabilitation: combining 3D convolution and LSTM with spatiotemporal attention by Fan Yang, Shiyu Li, Chang Sun, Xingjiang Li, Zhangbo Xiao (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-11646842.txt.
Description
Fan Yang and co-authors build a model that recognizes therapy exercises from video in a paper published in 2024. Rehab programs need fast, accurate checks of patient movement, often in messy home settings. Older methods struggle with the fine detail of such actions.
Their model joins three parts: a three-dimensional convolutional network for motion in space and time, a memory network for longer sequences, and an attention step that focuses on the most telling moments. On four public data sets it beat prior methods, reaching accuracy above ninety-six percent on one set. The authors aim to give therapists a tool for live feedback. Future work will test odd and extreme movements and more rehab settings.
Catalog record
| CREATOR | Fan Yang, Shiyu Li, Chang Sun, Xingjiang Li, Zhangbo Xiao |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |