GPRAR: Graph Convolutional Network based Pose Reconstruction and Action Recognition for Human Trajectory Prediction.
Manh HuynhGita AlaghbandPublished in: BMVC (2021)
Keyphrases
- action recognition
- human pose
- recognizing actions
- static images
- human activities
- human movements
- human actions
- bag of words
- activity recognition
- human detection
- action detection
- computer vision
- action classification
- body parts
- human object interactions
- motion capture data
- motion history images
- depth sensors
- view invariant
- recognition of human actions
- d objects
- pose estimation
- graph structure
- action recognition in videos
- video dataset
- depth cameras
- graph cuts
- partial occlusion
- three dimensional
- human body
- action primitives
- bag of features
- machine learning
- space time interest points