POOT: An efficient object tracking strategy based on short-term optimistic predictions for face-structured sensor networks.
Jenq-Muh HsuChao-Chun ChenChia Chi LiPublished in: Comput. Math. Appl. (2012)
Keyphrases
- sensor networks
- short term
- object tracking
- long term
- particle filter
- wireless sensor networks
- image sequences
- video sequences
- sensor data
- mean shift
- environmental monitoring
- video surveillance
- short term and long term
- kalman filter
- computer vision
- sensor nodes
- moving objects
- sparse representation
- underwater acoustic
- visual tracking
- long term memory
- communication cost
- communication bandwidth
- multiple cameras
- appearance model
- appearance modeling
- short term prediction
- energy consumption
- mobile sensor networks
- multi camera
- data streams
- load forecasting
- resource constrained
- routing protocol
- tracking accuracy
- face images
- sensor web
- face detection
- wind speed
- state space
- forecasting model
- feature extraction
- three dimensional
- surveillance system
- image processing