A comprehensive evaluation of state-of-the-art time-series deep learning models for activity-recognition in post-stroke rehabilitation assessment.
Issam BoukhennoufaXiaojun ZhaiVictor UttiJo JacksonKlaus D. McDonald-MaierPublished in: EMBC (2021)
Keyphrases
- activity recognition
- learning models
- comprehensive evaluation
- activities of daily living
- learning tasks
- human activity recognition
- human activities
- learning algorithm
- loss function
- machine learning
- transfer learning
- semi supervised learning
- machine learning algorithms
- action recognition
- learning problems
- conditional random fields
- classification models
- smart home
- smart environments
- daily activities
- ambient assisted living
- accelerometer data
- image segmentation
- higher order
- learning environment
- pairwise
- data sets