Automated Textual Descriptions for a Wide Range of Video Events with 48 Human Actions.
Patrick HanckmannKlamer SchutteGertjan J. BurghoutsPublished in: ECCV Workshops (1) (2012)
Keyphrases
- human actions
- human activities
- textual descriptions
- action recognition
- action sequences
- video sequences
- spatio temporal
- space time
- activity recognition
- human motion
- visual features
- recognition of human actions
- event detection
- recognizing human actions
- semantic concepts
- web images
- sensor data
- computer vision
- human computer interaction
- semantic representation
- event recognition
- key frames
- image sequences
- metadata
- semantic information
- user interface
- moving objects
- natural language
- multimedia