SYSTEM AND METHOD FOR AUTOMATICALLY DETECTING AND MARKING LOGICAL SCENES IN MEDIA CONTENT

A system including a scene detection engine (SDE) and a method for automatically detecting and marking logical scenes in media content using artificial intelligence are provided. The SDE detects boundaries of shots in media content. The SDE generates an image similarity matrix by extracting color features from a middle frame of each shot. The SDE generates an audio similarity matrix by extracting audio features of each shot. The SDE generates a resultant similarity matrix from the image and audio similarity matrices. The SDE executes a clustering algorithm using the resultant similarity matrix with reduced or no noise to generate an ordered sequence of shots that define a boundary of each scene of the media content. The SDE executes a linear traversal algorithm on the shots that define the boundary of each scene for correcting errors in the defined boundary of each scene.