Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: The Benefit of Target Expectation Maximization.
Guoliang KangLiang ZhengYan YanYi YangPublished in: ECCV (11) (2018)
Keyphrases
- expectation maximization
- em algorithm
- maximum likelihood
- mixture model
- target tracking
- probabilistic model
- parameter estimation
- target recognition
- focus of attention
- gaussian mixture
- dynamic time warping
- gaussian mixture model
- unsupervised learning
- information systems
- computer vision
- neural network
- image reconstruction
- visual attention
- bayesian framework
- multi agent
- face recognition
- clustering algorithm
- multiple targets
- mixture of gaussians