A multimodal Parkinson quantification by fusing eye and gait motion patterns, using covariance descriptors, from non-invasive computer vision.
John ArchilaAntoine ManzaneraFabio MartínezPublished in: Comput. Methods Programs Biomed. (2022)
Keyphrases
- motion patterns
- computer vision
- image sequences
- activity analysis
- spatio temporal
- space time
- motion model
- moving objects
- motion analysis
- appearance model
- multi modal
- image processing
- bayesian framework
- eye tracking
- video sequences
- humanoid robot
- hand gestures
- optical flow
- pose estimation
- object recognition
- object detection
- motion estimation
- gait recognition
- motion trajectories
- computer graphics
- object tracking
- face detection
- camera calibration
- machine learning
- mean shift
- signal processing
- motion capture
- three dimensional
- body motions
- em algorithm
- particle filter
- gesture recognition
- feature points
- higher order
- high dimensional
- feature space