Predicting Tongue Motion in Unlabeled Ultrasound Videos Using Convolutional Lstm Neural Networks.
Chaojie ZhaoPeng ZhangJian ZhuChengrui WuHuaimin WangKele XuPublished in: ICASSP (2019)
Keyphrases
- neural network
- ultrasound images
- motion analysis
- recurrent neural networks
- motion features
- moving camera
- recognition of human actions
- periodic motion
- static images
- video sequences
- motion model
- input video
- network architecture
- key frames
- successive frames
- optical flow
- image sequences
- motion capture data
- motion estimation
- dynamic scenes
- human motion
- traffic scenes
- camera motion
- motion segmentation
- video stabilization
- ultrasound probe
- video frames
- space time
- recognizing human actions
- temporal coherence
- neural network model
- computer aided
- moving objects
- motion parameters
- active learning
- motion cues
- dynamic textures
- motion capture
- deep learning
- semi supervised learning
- text classification
- adjacent frames
- spatio temporal
- training data
- video objects
- object motion
- back propagation
- video analysis
- motion patterns
- spatial and temporal
- video summarization
- labeled data
- feed forward
- unsupervised learning
- supervised learning
- video data
- video content
- learning algorithm