Automatic caption detection in video frames based on support vector machine

Authors:
Jianfeng Xu;Shaofa Li
Affiliations:
School of Computer Science, South China University of Technology, Guangdong, Guangzhou, China;School of Computer Science, South China University of Technology, Guangdong, Guangzhou, China
Venue:
ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II
Year:
2005

Citing 4
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Automatic text recognition for video indexing

MULTIMEDIA '96 Proceedings of the fourth ACM international conference on Multimedia
Finding text in images

DL '97 Proceedings of the second ACM international conference on Digital libraries
Video Skimming and Characterization through the Combination of Image and Language Understanding

CAIVD '98 Proceedings of the 1998 International Workshop on Content-Based Access of Image and Video Databases (CAIVD '98)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Video captions can be used to index large video archives in digital libraries. In this paper, an algorithm for detecting captions in video frames using support vector machine (SVM) is proposed. First, the input video frame is divided into square sub-blocks and a trained SVM is used to identify whether each sub-block is a caption block or not. Second, horizontal projection and vertical projection are performed to locate the candidate caption regions. Finally, false alarms are reduced by caption region verification. Experimental results show that the algorithm has a low missed rate and false alarm rate.