Title :
Video text extraction using temporal feature vectors
Author :
Tang, Xiaoou ; Luo, Bo ; Gao, Xinbo ; Pissaloux, Edwige ; Hangjiang Zhan
Author_Institution :
Dept. of Inf. Eng., Chinese Univ. of Hong Kong, Shatin, China
Abstract :
A new caption text extraction algorithm that takes full advantage of the temporal information in a video sequence is developed. By detecting the (dis)appearance of caption text in a video stream, we first identify the video segment that contains the same caption text. Then using the gray-level vector traced across the segment as the feature vector for a pixel point, we can clearly separate a caption pixel from a background pixel for the entire segment.
Keywords :
content-based retrieval; database indexing; feature extraction; image sequences; multimedia databases; video databases; video signal processing; background pixel; caption pixel; caption text extraction algorithm; gray-level vector; pixel point; temporal feature vectors; video segment; video sequence; video stream; video text extraction; Data mining; Feature extraction; Gunshot detection systems; Image edge detection; Indexing; Layout; Optical character recognition software; Robots; Video compression; Video sequences;
Conference_Titel :
Multimedia and Expo, 2002. ICME '02. Proceedings. 2002 IEEE International Conference on
Print_ISBN :
0-7803-7304-9
DOI :
10.1109/ICME.2002.1035724
