Color segmentation based depth image filtering

7Citations
Citations of this article
18Readers
Mendeley users who have this article in their library.
Get full text

Abstract

We present a novel enhancement method that addresses the problem of corrupted edge information in depth maps. Corrupted depth information manifests itself in zigzag edges instead of straight ones. We extract the depth information from an associated color stream and use this information to enhance the original depth map. Besides the visual results, a quantitative analysis is conducted to prove the capabilities of our approach. For this task, we introduce a new assessment technique which is based on measuring clustering similarity using the Rand index. © 2013 Springer-Verlag.

Cite

CITATION STYLE

APA

Schmeing, M., & Jiang, X. (2013). Color segmentation based depth image filtering. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7854 LNCS, pp. 68–77). https://doi.org/10.1007/978-3-642-40303-3_8

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free