METHOD FOR EXTRACTING FEATURES OF VIDEO USING ENCODERS BASED ON CONVOLUTIONAL NEURAL NETWORK

摘要

The present invention relates to feature extraction of a video. More specifically, provided is a method for extracting a video feature set using a plurality of encoders based on an artificial neural network, which comprises the steps of: extracting vectors from an input video using a plurality of artificial neural network classifiers; combining the vectors to reduce dimensions; and verifying whether the dimension-reduced vector is in a shape which meets a predetermined purpose in a corresponding application.