Extraction of human action elements with transition network of partial time series data modeled by Hidden Markov Model.
Kae DokiAkihiro ToriiSuguru MototaniYuki FunaboraShinji DokiKohjiro HashimotoPublished in: IECON (2016)
Keyphrases
- hidden markov models
- human actions
- sequential data
- action recognition
- speech recognition
- gesture recognition
- variable duration
- handwritten word recognition
- activity recognition
- hidden states
- viterbi algorithm
- conditional random fields
- human activities
- network structure
- space time
- human motion
- motion recognition
- spatio temporal
- video sequences
- continuous hidden markov models
- search engine
- generative model
- probabilistic model
- wireless sensor networks
- user interface
- image sequences
- computer vision