Multimodal Human Action Recognition Based on a Fusion of Dynamic Images Using CNN Descriptors.
Edwin Jonathan Escobedo CardenasGuillermo Cámara ChávezPublished in: SIBGRAPI (2018)
Keyphrases
- action recognition
- static images
- bag of features
- low level descriptors
- late fusion
- image retrieval
- bag of words
- input image
- action recognition in videos
- sift descriptors
- activity recognition
- mid level
- human actions
- image features
- human activities
- image classification
- human detection
- three dimensional
- video dataset
- object recognition
- object retrieval
- motion history images
- recognizing human actions
- image descriptors
- computer vision
- view invariant
- action classification
- motion capture data
- recognition of human actions
- image matching
- keypoints
- image registration
- global features
- color features
- feature descriptors
- body parts
- low level features
- mid level features
- space time
- machine learning