Title
Unsupervised Tracking With the Doubly Stochastic Dirichlet Process Mixture Model.
Abstract
We present an unsupervised tracking algorithm for human and car trajectory detection, using what is called the temporal doubly stochastic Dirichlet process (TDSDP) mixture model. The TDSDP captures the global dependence and the variation of human crowds and cars in temporal domains without the Markov assumption, making it particularly suitable for long-term tracking. Moreover, TDSDP prior can esti...
Year
DOI
Venue
2016
10.1109/TITS.2016.2518212
IEEE Transactions on Intelligent Transportation Systems
Keywords
Field
DocType
Hidden Markov models,Trajectory,Mixture models,Vehicles,Computational modeling,Feature extraction,Markov processes
Latent Dirichlet allocation,Dirichlet process,Markov process,Markov property,Pattern recognition,Video tracking,Cox process,Artificial intelligence,Hidden Markov model,Mixture model,Mathematics
Journal
Volume
Issue
ISSN
17
9
1524-9050
Citations 
PageRank 
References 
3
0.38
12
Authors
3
Name
Order
Citations
PageRank
Sun Xing13310.94
Nelson H. C. Yung227420.99
edmund y lam368369.87