METHOD AND SYSTEM FOR MULTI-VIEW OBJECT SEGMENTATION USING DEPTH IMAGE

摘要

The present invention relates to a method and system for separating a multi-view object using a depth image, comprising: acquiring a depth image corresponding to each of the target images based on the depth image, A system setup step of generating an inter-viewpoint correspondence area based on the system setup step; And generating a single image energy term based on the color information of each of the target images, further generating an inter-view energy term reflecting the difference between the inter-view color and the depth information based on the inter-view correspondence region, And an object separating step of separating the objects included in each of the target images through optimization of the Markov random field energy.