| R. Lienhart, Indexing and retrieval of digital video sequences based on automatic text recognition, in: Proc. 4th ACM Internat. Multimedia Conf., 1996. |
....provide information relating to where, when, and who the reported events pertain to. Such information is extremely useful for video indexing. A number of algorithms for detecting embedded text (captions) from still images and video sequences have been published in recent years [AD99, JY98, Lie96, SDB98, SK97, WMR99] All these algorithms operate on raw pixel (uncompressed) data. Compressed video needs to be decompressed before these algorithms can be applied. There has been very little e#ort in utilising features in the compressed domain for text detection. Yeo and Liu [YL95] proposed ....
R. Lienhart. Indexing and retrieval of digital video sequences based on automatic text recognition. Technical Report 6/96, University of Mannheim, 1996.
..... Characters contrast strongly with their background since text is designed to be read easily. The same characters appear in multiple consecutive frames. Characters appear in clusters of sharp edges confined in a horizontal rectangular region. Based on these characteristics, Lienhart [4] described a method of detecting text regions using color based segmentation. Each frame is first segmented into several regions according to the average color attribute. The size Scene Analysis of Video Sequences in the MPEG Domain Lifang Gu CSIRO Mathematical and Information Sciences GPO Box ....
R. Lienhart, Indexing and retrieval of digital video sequences based on automatic text recognition, Technical Report 6/96, University of Mannheim, 1996.
No context found.
R. Lienhart, Indexing and retrieval of digital video sequences based on automatic text recognition, in: Proc. 4th ACM Internat. Multimedia Conf., 1996.
No context found.
R. Lienhart, Indexing and retrieval of digital video sequences based on automatic text recognition, in: Proc. 4th ACM Internat. Multimedia Conf., 1996.
No context found.
R. Lienhart, F. Stuber (1996) Indexing and retrieval of digital video sequences based on automatic text recognition. In: Proc. ACM Int. Multimedia Conf. & Exhibition, pp 419--420
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC