PipeSLAM: Simultaneous localisation and mapping in feature sparse water pipes using the Rao-Blackwellised particle filter.
Ke MaMichele M. SchirruAli Hassan ZahraeeRob Dwyer-JoyceJoby B. BoxallTony J. DoddRichard CollinsSean R. AndersonPublished in: AIM (2017)
Keyphrases
- particle filter
- object tracking
- particle filtering
- visual tracking
- state estimation
- kalman filter
- water flow
- multiple hypotheses
- water distribution
- mean shift
- monte carlo
- robust visual tracking
- state space
- motion model
- likelihood function
- observation model
- appearance model
- data association
- robust tracking
- rao blackwellized particle filter
- face tracking
- color histogram
- sequential monte carlo
- proposal distribution
- sparse representation
- image features
- multiple hypothesis
- multiple object tracking
- importance sampling
- extended kalman filter
- multiple objects
- tracking scheme
- high dimensional
- reinforcement learning