Poster abstract: Gesture recognition via continuous maximum entropy training on accelerometer data.
Alan L. LiuJun YangPéter Pál BodaPublished in: IPSN (2009)
Keyphrases
- maximum entropy
- gesture recognition
- accelerometer data
- iterative scaling
- human computer interaction
- maximum entropy principle
- hidden markov models
- sign language recognition
- hand gestures
- conditional random fields
- sign language
- activity recognition
- human machine interface
- behavior recognition
- training set
- principle of maximum entropy
- minimum cross entropy
- transformation based learning
- sign recognition
- supervised learning
- expert systems
- object recognition
- reinforcement learning