Title :
Contour-traversal based algorithm for insertion of virtual advertisements in videos
Author :
Hegde, Akshay ; Kumar, Deepa Anil ; Srinivasa, Gowri
Author_Institution :
Dept. of Comput. Sci. & Eng., PES Inst. of Technol., Bangalore, India
Abstract :
This paper presents a method for the automated insertion of virtual advertisements (logos, messages, etc.) into videos. The method consists of detecting regular polygons from the edge maps of frames in a video. This is followed by the selection of an optimal candidate slot into which the desired advertisement is inserted. Experimental results demonstrate that the contour-tracing based rectangle detection algorithm presented in this paper outperforms its precedents such as the Hough Transform and ant-based polygon detection for the purposes of identifying appropriate slots for introducing content in videos.
Keywords :
Hough transforms; advertising; computational geometry; edge detection; object detection; video signal processing; Hough transform; ant-based polygon detection; contour-tracing based rectangle detection algorithm; contour-traversal based algorithm; edge maps; regular polygon detection; videos; virtual advertisements insertion; Computers; Gray-scale; Image edge detection; Indexes; Shape; Transforms; Videos; contour traversal; edge-map; shape detection; virtual advertisements;
Conference_Titel :
Communication, Information & Computing Technology (ICCICT), 2012 International Conference on
Conference_Location :
Mumbai
Print_ISBN :
978-1-4577-2077-2
DOI :
10.1109/ICCICT.2012.6398213