Video Indexing and Retrieval System
A video indexing and retrieval system (VIRS) focuses on the efficient and reliable retrieval of video contents on the web. Since the MPEG compression techniques have been very popular now, so incorporating the compression technique in content based search to index and retrieve compressed data efficiently and quickly has been an important area of research. VIRS consists of four components: Shot Detection, Shot Representation, Color Feature Representation and Motion Feature Representation. The motion based classifier is used to capture the moving images and further color database (which have been quantized on the basis of ten most commonly used colors) has been used for testing the accuracy of system. The user is allowed to fetch the query either by typing or by sketching. The graphics user interface displays the results of query using a visual color chart of 10 quantized colors, a visual shape chart of common shapes, a visual texture chart of common textures with known feature vectors and a menu of video type buttons.