Duration Dependent Codebooks for Change Detection

Brandon Mayer and Joseph Mundy

In Proceedings British Machine Vision Conference 2014


This paper describes a supervised system for pixel-level change detection for fixed, monocular surveillance cameras. Per-pixel intensity sequences are modeled by a class of Hidden Semi-Markov Models, Duration Dependent Hidden Markov Models (DDHMMs), to accurately account for stocastically periodic phenomena prevalent in real-world video. The per-pixel DDHMMs are used to assign discrete state labels to pixel intensity sequences which summarize the appearance and temporal statistics of the observations. State assignments are then used as a features for constructing per-pixel code books during a training phase to identify changes of interest in new video. The per-pixel intensity model is validated by showing superior predictive performance to pixel representations commonly used in change detection applications. A new data set is presented which contain dynamic, periodic backgrounds with larger time scale variability than previous data sets and the proposed method is compared to state-of-the-art change detection methods using the new videos.


Poster Session


Extended Abstract (PDF, 1 page, 1.9M)
Paper (PDF, 10 pages, 1.8M)
Bibtex File


Brandon Mayer, and Joseph Mundy. Duration Dependent Codebooks for Change Detection. Proceedings of the British Machine Vision Conference. BMVA Press, September 2014.


	title = {Duration Dependent Codebooks for Change Detection},
	author = {Mayer, Brandon and Mundy, Joseph},
	year = {2014},
	booktitle = {Proceedings of the British Machine Vision Conference},
	publisher = {BMVA Press},
	editors = {Valstar, Michel and French, Andrew and Pridmore, Tony}
	doi = { http://dx.doi.org/10.5244/C.28.126 }