2D and 3D video scene text classification

Xu, J. and Shivakumara, P. and Lu, T. and Tan, C.L. (2014) 2D and 3D video scene text classification. In: International Conference on Pattern Recognition (ICPR) , 24-28 Aug 2014, Stockholm, Sweden. (Submitted)

2d_and_3d_video_scene_text.pdf - Submitted Version

Download (2MB)


Text detection and recognition is a challenging problem in document analysis due 10 the presence of the unpredictable nature of video texts, such as the variations of orientation, font and size, illumination effects, and even different 20/30 text shadows. In this paper, we propose a novel horizontal and vertical symmetry feature by calculating the gradient direction and the gradient magnitude of each text candidate, which results in Potential Text Candidates (PTCs) after applying the k-means clustering algorithm on the gradient image of each input frame to verify PTC , we explore temporal information of video by proposing an iterative process that continuously verifies the PTCs of the first frame and the successive frames, until the process meets the converging criterion. This outputs Stable Potential Text Candidates (SPTCs). For each , PTC, the method obtains text representatives with the help of the edge image of the input frame. Then for each text representative, we divide it into four quadrants and check a new Mutual Nearest Neighbor Symmetry (MNNS) based on the dominant stroke width distances of the four quadrants. A voting method is finally proposed to clasify each text block as either 2D or 3D by counting the text representatives that satisfy MNNS. Experimental results on clasifying 2D and 3D text images are promising, and the result re further validated by text detection and recognition before clasification and after clasification with the exiting methods, respectively.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Video text frames, Horizontal and vertical symmetry, Video potential text candidates, Dominant potential text candidates, 2D and 3D text video classification
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Computer Science & Information Technology
Depositing User: Mr. Mohd Samsul Ismail
Date Deposited: 24 Mar 2015 06:14
Last Modified: 24 Mar 2015 06:14
URI: http://eprints.um.edu.my/id/eprint/13101

Actions (login required)

View Item View Item