Discrimination of abandoned and stolen object based on active contours

Abstract

Esta es la versión de autor de la comunicación de congreso publicada en: This is an author produced version of a paper published in:

Exploiting Scene Cues for Dropped Object Detection

“…This problem can be alleviated by exploiting object detectors for known elements in the scene (such as humans). Surprisingly, and despite recent advances, existing dropped and abandoned object detection methods aiming at realistic surveillance scenarios (Tian et al, 2011;Caro Campos et al, 2011;Fan and Pankanti, 2011;Fan and Pankanti, 2012) do not attempt to use state-of-the-art object detection methods (Dalal and Triggs, 2005;Felzenszwalb et al, 2010), which could help in reducing false alarms caused by such known objects.…”

“…Substantial research effort has been recently devoted to providing robust solutions to dropped and abandoned object detection (Smith et al, 2006;Caro Campos et al, 2011;Tian et al, 2011;Fan and Pankanti, 2012). Because dropped objects can belong to different classes, foreground detection and adaptive background modeling are central elements in most of the existing approaches.…”

“…Liao et al (Liao et al, 2008) propose a foreground-mask sampling approach consisting in the intersection of foreground masks within a period of time. Campos et al (Caro Campos et al, 2011) rely on active contours obtained from foreground masks in order to detect abandoned and stolen objects.…”

Patch-based local histograms and contour estimation for static foreground classification

“…The Sobel [5,[7][8][9][10][11][12] and Canny [5,[13][14][15][16] operators have been employed with greater frequency for this purpose than the SUSAN edge detector [17]. While edge energy was employed in [17][18][19].…”

“…Color histogram comparison [15,[22][23][24] is another intuitive manner to discriminate abandoned and removed objects. Researchers compare the color distributions of the interior and exterior neighborhood of foreground blobs.…”