Long and Short Memory Balancing in Visual Co-Tracking Using Q-Learning.
Kourosh MeshgiMaryam Sadat MirzaeiShigeyuki ObaPublished in: ICIP (2019)
Keyphrases
- reinforcement learning
- long sequences
- cooperative
- real time
- memory usage
- visual information
- visual perception
- function approximation
- visual tracking
- particle filter
- visual features
- short term memory
- memory space
- state space
- image sequences
- motion model
- multi agent
- appearance model
- memory requirements
- high level
- particle filtering
- learning algorithm
- moving target
- kalman filter
- potential field
- action selection
- smooth pursuit
- visual cues
- visual data
- mean shift
- optimal policy
- low level
- video sequences
- neural network