Active appearance models for photorealistic visual speech synthesis.
Wesley MattheysesLukas LataczWerner VerhelstPublished in: INTERSPEECH (2010)
Keyphrases
- photorealistic
- active appearance models
- speech synthesis
- speech recognition
- virtual reality
- computer graphics
- highly accurate
- range images
- text to speech
- hand held
- appearance model
- complex scenes
- statistical model
- image space
- vision system
- d scene
- image acquisition
- low resolution
- point cloud
- shape model
- human faces
- facial images
- generative model
- pose estimation
- deformable models
- computer vision
- feature points
- training set
- high quality
- similarity measure
- image processing