Accelerometer-based activity recognition on a mobile phone using cepstral features and quantized gmms.
Jussi LeppänenAntti J. EronenPublished in: ICASSP (2013)
Keyphrases
- activity recognition
- mobile phone
- cepstral features
- gaussian mixture model
- human activities
- human activity recognition
- feature extraction
- daily activities
- mobile devices
- action recognition
- mobile users
- smart phones
- sensory data
- mobile phone users
- smart home
- mobile applications
- accelerometer data
- human actions
- smart environments
- transfer learning
- wearable sensors
- activities of daily living
- computer vision
- training set
- text entry
- image processing