Continuous inaudible recognition of Japanese vowels using features detected at lips shape peaks based on surface electromyography.
Nao KurogiHidetoshi NagaiTeigo NakamuraPublished in: ICSAI (2016)
Keyphrases
- feature extraction
- appearance features
- image features
- surface features
- speech recognition systems
- feature set
- recognition rate
- statistical shape model
- extended gaussian image
- lip reading
- geometrical features
- shape recovery
- object recognition
- shape descriptors
- speech recognition
- d mesh
- three dimensional
- object representation
- shape properties
- action recognition
- face recognition
- free form surfaces
- d objects
- hand shape
- deformable surface
- global shape
- surface shape
- surface patches
- shape features