Predicting Tongue Motion in Unlabeled Ultrasound Video Using 3D Convolutional Neural Networks.
Chengrui WuShicheng ChenGuorui ShengPierre RousselBruce DenbyPublished in: ICASSP (2018)
Keyphrases
- convolutional neural networks
- ultrasound images
- motion analysis
- space time
- motion features
- object motion
- video sequences
- key frames
- temporal filtering
- spatial and temporal
- video data
- moving camera
- temporal continuity
- input video
- dynamic scenes
- static images
- motion estimation
- video streams
- multimedia
- dynamic textures
- visual cues
- low frame rate
- surveillance videos
- shot change detection
- video content
- visual data
- successive frames
- motion trajectories
- motion model
- ultrasound probe
- temporal consistency
- video frames
- video summarization
- camera motion
- convolutional network
- video scene
- camera movement
- video tracking
- motion capture data
- video analysis
- optical flow
- video shots
- image sequences
- temporal coherence
- active learning
- healthy volunteers
- appearance features
- video objects
- computer aided
- motion capture
- video surveillance
- semi supervised learning
- human motion
- moving objects
- layered representation
- motion segmentation
- active contour model