Approach for retrieving and mining video clips

Abstract EN

Multimedia (include video, images, audio and text media) is characterized by its high dimensionality, which makes information retrieval and data mining even more challenging.

This research proposesa method to build an indexes database for huge collection of video clips, to make the video retrieval and mining much more efficient and perfectthat by considering similarity in both text of sound and features of frames.

The proposed method has the following steps: First, isolates video motion from sound in the video clips.

Second, converts the sound to text and index the result with database.

Third converts video motion to shots, then select the master frame for each oneand extracts the feature vector for them such as color, texture, shape and others and finally index the result with database.

Fourth,combines the two resulted indexed database (Second and Third steps)into one database and make it the final and standard for both retrieval and mining.