|Conference/Proceedings||Visual Communications and Image Processing (VCIP), IS&T/SPIE's Electronic Imaging 2009|
|Address||San Jose, CA, USA|
|Author(s)||Mustafa Karaman, Lutz Goldmann, Thomas Sikora|
|Title||Improving object segmentation by reﬂection detection and removal|
|Abstract||For object analysis in videos such as in video surveillance systems, the preliminary segmentation step is very important. Many segmentation methods using static camera have been proposed in the last decade, but they all suﬀer in occurrance of object reﬂection especially on the ground, i.e. reﬂected regions are also segmented as foregrounds. We present a new method which detects the border between the real object and its reﬂection. Experiments show that an outstanding improvement|
of segmentation results are obtained by removing the reﬂection part of the over-segmented objects.
|Key words||object segmentation, reflection detection, background subtraction|