METHOD AND SYSTEM FOR EXTRACTING HIGH-LEVEL FEATURES FROM LOW-LEVEL FEATURES OF MULTIMEDIA CONTENT

摘要

PROBLEM TO BE SOLVED: To extract features from a video including a sequence of frames. SOLUTION: A plurality of high-level features 109 are extracted from a video 101 including a sequence of frames and plurality of low-level features are extracted 111-113 from each frame of the video. Each frame of the video is labeled 120 according to the extracted low-level features to generate a plurality of sequences of labels. Each sequence of labels is associated with one of the extracted low-level features. The plurality of sequences of labels are analyzed 130 using machine learning techniques to extract high-level features of the video.