Identifying key frame with the use of a thinner based on population groups analysis

摘要

A method for identifying a set of key video frame of a video sequence, which comprises extracting feature vectors for each video frame and applying an algorithm based on a thin population groups, in order to determine the feature vector for a video frame as a thin based on population groups combination of the feature vectors for the other video frame to represent. Weighting coefficients, which leads to the thinner based on population groups combination, are analyzed in order to video frame - cluster of temporally coherent similar video frame to determine. A set of key video frame is based on the particular video frame - cluster determined.