Human Motion Generation by Stochastic Conditioning of Deep Recurrent Networks On Pose Manifolds.
Himanshu BuckchashBalasubramanian RamanPublished in: ICIP (2020)
Keyphrases
- human motion
- recurrent networks
- human pose
- body configuration
- human body
- motion capture
- recurrent neural networks
- articulated motion
- pose estimation
- video sequences
- body parts
- articulated objects
- biologically inspired
- motion sequences
- human motion tracking
- image sequences
- motion recognition
- optical motion capture
- spatio temporal
- feed forward
- gait recognition
- human movement
- neural network
- human actions
- motion synthesis
- manifold learning
- human movements
- human pose estimation
- body pose
- humanoid robot
- articulated body
- markerless
- body posture
- d objects
- human motion analysis
- visual data
- euclidean space
- dimensionality reduction
- feature points
- depth images
- motion capture data
- geometric structure