An Annotated-Objects Assist Method for Extraction of Ordinary-Objects in a Video Content Generation Support System

摘要

In our previous work, we proposed a semantic video content generation support system, based on an interactive approach that maps low-level features to semantic concepts. By consulting an ontological semantic object model database, the main semantic objects in key frames of each video shot can be extracted and annotated based on similarities in color, area, and position of each region. This system has high potential for use in object-based interactive multimedia applications. This paper extends our previous works, first by showing system evaluation results, and then by proposing an object extraction assist method that aims at ordinary objects by combining low-level features and high-level concepts of annotated objects.