Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and Detection.
Mohammadreza ZolfaghariGabriel L. OliveiraNima SedaghatThomas BroxPublished in: CoRR (2017)
Keyphrases
- feature vectors
- action classification
- static images
- multi stream
- loop closure
- action recognition
- human pose
- feature extraction
- image sequences
- human detection
- human actions
- space time
- pose estimation
- object detection
- d objects
- hidden markov models
- video sequences
- partial occlusion
- audio visual
- bag of words
- human motion
- moving objects
- human body
- body parts
- visual data
- computer vision