Automatic video annotation framework using concept detectors

• Main Authors: Chamasemani, Fereshteh Falah, Affendey, Lilly Suriani, Mustapha, Norwati, Khalid, Fatimah
• Format: Article
• Language: English
• Published: Asian Network for Scientific Information 2015
• Online Access: http://psasir.upm.edu.my/id/eprint/43900/1/Automatic%20video%20annotation%20framework%20using%20concept%20detectors.pdf; http://psasir.upm.edu.my/id/eprint/43900/; http://www.scialert.net/qredirect.php?doi=jas.2015.256.263&linkid=pdf

Automatic video annotation has received a great deal of attention from researchers working on video retrieval. This study presents a novel automatic video annotation framework to enhance the annotation accuracy and reduce the processing time in large-scale video data by utilizing semantic concepts. The proposed framework consists of three main modules i.e., pre-processing, video analysis and annotation module. The framework support an efficient search and retrieval for any video content analysis and video archive applications. The experimental results on widely used TRECVID dataset using concepts of Columbia374 demonstrate the effectiveness of the proposed framework in assigning appropriate and semantically representative annotations for any new video.