APPARATUS AND METHOD FOR SEGMENTING OBJECT BY USING GAUSSIAN MIXTURE MODEL AND TOTAL VARIATION TECHNIQUE

摘要

Disclosed are an apparatus and a method for segmenting an object by using a Gaussian mixture model and a total variation technique. According to the present disclosure, the method for segmenting an object includes the steps of: receiving an input image; receiving a user input indicating information on at least one of a foreground region and a background region included in the input image; generating at least one among a foreground pixel list and a background pixel list by using the received user input; calculating Gaussian distribution of at least one pixel which makes up the input image by using at least one of the generated foreground pixel list and the background pixel list; and determining whether the at least one pixel is a foreground pixel or a background pixel by using the calculated Gaussian distribution.;COPYRIGHT KIPO 2019