Article

Article title DEVELOPMENT AND RESEARCH ALGORITHM FOR COMPUTING THE DEPTH MAPS OF STEREO IMAGES
Authors V.V. Voronin
Section SECTION I. METHODS IMAGE AND SIGNAL PROCESSING
Month, Year 11, 2013 @en
Index UDC 004.932.2
DOI
Abstract The problems of computing the depth map of the scene based of the modified method Depth Transfer. The proposed approach is based on finding similar images in the database, which is formed by the depth map, with subsequent post-processing to reduce the error of the image conversion from 2D to 3D. The basic idea of the method is that similar scenes have approximately the same distribution of the depth map. For more accurate estimation and interpolation with the use of global optimization using several similar candidates for the image. The proposed post-processing algorithm consists of the depth map of the following steps: filtering the image in RGB space Gaussian filter, clustering of images using k-means, adaptive median filtering of the depth map within each cluster, a color image, the removal of small areas of closed "holes" a depth map. The effectiveness of the new approach is shown in several examples that demonstrate the results of recovery of the depth map on the test images with different geometric features.

Download PDF

Keywords Depth map; filtering; contour analysis; descriptors; stereo image.
References 1. Форсайт Д.А., Понс Ж. Компьютерное зрение. Современный подход: Пер. с англ. – М.: Издательский дом «Вильямс», 2004. – 928 с.
2. Torralba A., Oliva A. Depth Estimation from Image Structure // IEEE Transactions on Pattern Analysis and Machine Intelligence. – 2002. – Vol. 24, Issue 9. – P. 1226-1238.
3. Kevin Karsch, Ce Liu, Sing Bing Kang. Depth Extraction from Video Using Non-parametric Sampling // Lecture Notes in Computer Science. – 2012. – Vol. 7576. – P. 775-788.
4. Oliva A., Torralba A. Modeling the shape of the scene: A holistic representation of the spatial envelope // IJCV. – 2001. – № 42. – P. 145-175.
5. Liu C., Yuen J., Torralba A. SIFT Flow: Dense correspondence across scenes and its applications // IEEE TPAMI. – 2011. – № 33. – P. 978-994.
6. Konrad J., Wang M., Ishwar P. 2D-to-3D image conversion by learning depth from examples // In: 3DCINE. – 2012. – P. 75-88.
7. Франц В.А., Левина О.М., Воронин В.В., Кожин Р.А. Первичная обработка карты глубины изображения. Успехи современной радиоэлектроники // Радиотехника. – 2013. – № 6. – С. 39-43.

Comments are closed.