CausalBatch: solving complexity/performance tradeoffs for deep convolutional and LSTM networks for wearable activity recognition.
Lloyd PellattDaniel RoggenPublished in: UbiComp/ISWC Adjunct (2020)
Keyphrases
- activity recognition
- accelerometer data
- human activity recognition
- action recognition
- human activities
- wearable sensors
- daily activities
- deep learning
- daily life
- activities of daily living
- visual surveillance
- smart home
- deep belief networks
- neural network
- sensory data
- smart environments
- human actions
- transfer learning
- human computer interaction
- unsupervised learning
- supervised learning
- active learning
- learning algorithm
- data sets
- eye movement analysis