Abstract
An efficient stereo matching algorithm for computing stable disparity map sequence from video footage is presented. The algorithm is based on both the spatial and temporal consistency in the stereo sequences, and high quality disparity maps are achieved. Weber local descriptors (WLD) are extracted for each color channel from current stereo pairs, and the raw matching costs between the images are initialized by WLD. Orthogonal integral image (OII) technique along with minimum spanning tree (MST) is used to aggregate the similar pixels and preserve disparity edges adaptively. MST takes place of the process of voting support regions in OII technique and provides a specific support region for each pixel. The nodes of MST are all the image pixels, and the weight of edges are absolute difference between the nearest neighboring pixels. It’s a global method, and can achieve more accurate disparity maps than traditional OII technique. Three-frame subtraction is used to determine the temporal consistency between adjacent frames. The motion region is extracted and the disparity map of motive region is renewed. The disparity of current frame with the renewed disparity and the one for last frame is confirmed. The proposed approach has been tested on the real stereo sequences, and the results are satisfactory.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.