Self-Supervised Joint Encoding of Motion and Appearance for First Person Action Recognition.
Mirco PlanamenteAndrea BottinoBarbara CaputoPublished in: ICPR (2020)
Keyphrases
- action recognition
- static images
- recognizing human actions
- recognition of human actions
- action recognition in videos
- human movements
- human actions
- motion features
- bag of words
- view invariant
- action primitives
- activity recognition
- computer vision
- motion capture data
- human detection
- action classification
- mid level
- motion history images
- object motion
- bag of features
- image sequences
- body parts
- human pose
- human activities
- recognizing actions
- optical flow
- space time
- motion estimation
- machine learning
- depth sensors
- action detection
- appearance model
- appearance information
- bag of visual words
- object segmentation