Learning attentional policies for tracking and recognition in video with deep networks.
Loris BazzaniNando de FreitasHugo LarochelleVittorio MurinoJo-Anne TingPublished in: ICML (2011)
Keyphrases
- real time
- learning process
- reinforcement learning
- learning algorithm
- visual recognition
- recognition rate
- neural network
- face detection and tracking
- unsupervised learning
- state space
- active learning
- motion model
- human activities
- video content
- video retrieval
- moving camera
- deep learning
- visual learning
- connectionist networks