Advertisement
Published in:
01-12-2013 | Original Article
Automatic non-interactive matting of dynamic targets in complex video using Graph Cuts
Published in: International Journal of Machine Learning and Cybernetics | Issue 6/2013
Log inActivate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
Extracting moving targets from video automatically and accurately is of great significance in the field of intelligent transport. This paper is devoted to the video object matting problem which is based on solving an optimization problem. The matting problem is severely ill-posed and generally cannot be solved without any extra constraints. In contrast to previous efforts, we develop a new method of automatic non-interactive matting of dynamic objects in complex video. Combining the Orthogonal Gaussian–Hermite Moments algorithm and Graph Cuts, highly stabile scribbles become automatic generation without interaction in the method and are used as the extra constraints in calculating the minimum energy function. The real-time of moving object matting may be improved obviously. Experiments demonstrate that the proposed method can effectively extract moving targets from video in a real-time way. And the method can be widely applied and suited to the extraction of complex video images due to the robustness of these scribbles.