Multimodal Fusion of Appearance Features, Optical Flow and Accelerometer Data for Speech Detection.
Panagiotis GiannakerisStefanos VrochidisIoannis KompatsiarisPublished in: MediaEval (2019)
Keyphrases
- appearance features
- optical flow
- multimodal fusion
- accelerometer data
- audio visual
- multimodal interfaces
- activity recognition
- image sequences
- detection method
- conditional random fields
- multi modal
- machine learning
- contextual information
- active shape model
- human computer interaction
- high robustness
- decision trees
- computer vision