Abstract | ||
---|---|---|
This paper addresses the problem of foreground extrac- tion using active illumination and graph-cut optimization. Our approach starts by detecting image regions that are likely to belong to foreground objects. These regions are constituted by pixels where the difference in luminance for two differently illuminated images is large. The fore- ground objects are segmented by graph-cut optimization us- ing those regions as a seed and using a energy function based on probability distributions derived from both input images and their difference. Several light sources and dif- ferent illumination schemes can be used to mark the fore- ground. Our method has only two scalar parameters which can be set once for a wide variety of scenes. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1109/SIBGRAPI.2006.5 | SIBGRAPI |
Keywords | Field | DocType |
feature extraction,graph theory,image segmentation,optimisation,statistical distributions,active illumination,energy function,foreground extraction,foreground object segmentation,graph-cut optimization,probability distribution | Graph theory,Cut,Computer vision,Computer science,Scalar (physics),Feature extraction,Image segmentation,Probability distribution,Pixel,Artificial intelligence,Luminance | Conference |
ISSN | Citations | PageRank |
1530-1834 | 2 | 0.42 |
References | Authors | |
20 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Asla Medeiros Sá | 1 | 64 | 8.43 |
Marcelo Bernardes Vieira | 2 | 50 | 15.30 |
Anselmo Antunes Montenegro | 3 | 18 | 6.60 |
Paulo Cezar Pinto Carvalho | 4 | 74 | 12.29 |
Luiz Velho | 5 | 1162 | 120.74 |