Image Recognition Based on Convolutional Neural Networks Using Features Generated from Separable Lattice Hidden Markov Models.
Takayuki KasugaiYoshinari TsuzukiKei SawadaKei HashimotoKeiichiro OuraYoshihiko NankakuKeiichi TokudaPublished in: APSIPA (2018)
Keyphrases
- hidden markov models
- image recognition
- speech recognition
- convolutional neural networks
- markov models
- image classification
- sequence classification
- continuous hidden markov models
- pattern recognition
- feature space
- gesture recognition
- markov model
- discriminative classifiers
- feature vectors
- feature extraction
- conditional random fields
- face recognition
- multiresolution
- object recognition
- sequence data
- viterbi algorithm
- hidden state
- visual speech recognition
- computer vision