Human Action Adverb Recognition: ADHA Dataset and a Three-Stream Hybrid Model.
Bo PangKaiwen ZhaCewu LuPublished in: CVPR Workshops (2018)
Keyphrases
- human actions
- hybrid model
- action recognition
- human activities
- activity recognition
- motion recognition
- recognizing actions
- human motion
- artificial neural networks
- action classification
- spatio temporal
- hybrid models
- support vector regression
- bag of words
- space time
- motion history images
- data streams
- video sequences
- human object interactions
- computer vision
- back propagation neural network
- gait recognition
- view invariant
- pattern recognition
- recognizing human actions
- visual features
- object recognition
- action recognition in videos
- sensor data
- human pose
- data sets
- support vector machine svm
- support vector machine
- feature extraction
- image sequences
- high level
- image processing
- learning algorithm
- neural network