DocumentCode :
617407
Title :
Keeping count: Leveraging temporal context to count heavily overlapping objects
Author :
Fiaschi, Luca ; Konstantin, Georg ; Afonso, Bruno ; Zlatic, Marta ; Hamprecht, Fred A.
Author_Institution :
HCI, IWR, Heidelberg, Germany
fYear :
2013
fDate :
7-11 April 2013
Firstpage :
656
Lastpage :
659
Abstract :
When tracking and segmenting multiple objects under heavy occlusion, a large class of algorithms can greatly benefit from a preprocessing that reliably assesses the number of individuals in each cluster. This is a difficult task when relying on local information only, due to scarcity of training examples and lack of strongly predictive features. In this paper, we develop a deterministic graphical model to address the problem of counting the number of objects in each foreground region as global inference across the entire video sequence. We show that global inference improves over local predictions, and is able to produce an accurate and coherent output within an useful runtime.
Keywords :
biological techniques; biology computing; image segmentation; image sequences; object tracking; graphical model; leveraging temporal context; multiple object segmentation; object tracking; video sequence; Computational modeling; Graphical models; Probabilistic logic; Random variables; Reliability; Standards; Training; constraint satisfaction; deterministic higher order potential; spatio-temporal segmentation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Imaging (ISBI), 2013 IEEE 10th International Symposium on
Conference_Location :
San Francisco, CA
ISSN :
1945-7928
Print_ISBN :
978-1-4673-6456-0
Type :
conf
DOI :
10.1109/ISBI.2013.6556560
Filename :
6556560
Link To Document :
بازگشت