Convolutional nets and watershed cuts for real-time semantic Labeling of RGBD videos.
Camille CouprieClément FarabetLaurent NajmanYann LeCunPublished in: J. Mach. Learn. Res. (2014)
Keyphrases
- real time
- image segmentation
- video sequences
- surveillance videos
- video event
- high definition
- sports video
- vision system
- semantic information
- natural language
- semantic web
- semantic annotation
- video images
- semantic search
- semantic concepts
- video surveillance
- high level
- unsupervised learning
- control system
- active learning
- multiscale
- watershed segmentation
- event recognition
- gradual transitions
- video database
- video analysis
- semantic similarity
- video content
- human activities