Efficient Learning of the Parameters of Non-Linear Models using Differentiable Resampling in Particle Filters.
Conor RosatoPaul R. HorridgeThomas B. SchönSimon MaskellPublished in: CoRR (2021)
Keyphrases
- efficient learning
- particle filter
- parameter estimation
- particle filtering
- estimation problems
- visual tracking
- object tracking
- monte carlo
- likelihood function
- maximum likelihood
- observation model
- importance sampling
- mean shift
- probabilistic model
- software systems
- log linear models
- dynamical models
- markov chain monte carlo
- kalman filter
- robust tracking
- graph cuts
- multiple object tracking