METHOD FOR IMPROVING QUALITY OF 2D-TO-3D AUTOMATIC CONVERSION BY USING MACHINE LEARNING TECHNIQUES

摘要

A method for improving the quality of 2D-to-3D automatic conversion by using machine learning techniques comprises that: receiving new 2D image from user input (S101); recognizing multiple objects in the new 2D image by machine learning algorithm (S102); obtaining the hierarchical relations among objects according to the features of the said objects (S103); improving the quality of 3D image based on the hierarchical relations (S104). The method can provide a more distinct classification of objects in a 2D image, thereby improving the 2D-to-3D conversion.