Tampere University of Technology

TUTCRIS Research Portal

Depth map occlusion filling and scene reconstruction using modified exemplar-based inpainting

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Details

Original languageEnglish
Title of host publicationImage Processing: Algorithms and Systems XIII
PublisherSPIE
ISBN (Print)9781628414899
DOIs
Publication statusPublished - 2015
Publication typeA4 Article in a conference publication
EventIS&T/SPIE Electronic Imaging / Image Processing: Algorithms and Systems -
Duration: 1 Jan 1900 → …

Publication series

NameSPIE Conference Proceedings
Volume9399

Conference

ConferenceIS&T/SPIE Electronic Imaging / Image Processing: Algorithms and Systems
Period1/01/00 → …

Abstract

RGB-D sensors are relatively inexpensive and are commercially available off-the-shelf. However, owing to their low complexity, there are several artifacts that one encounters in the depth map like holes, mis-alignment between the depth and color image and lack of sharp object boundaries in the depth map. Depth map generated by Kinect cameras also contain a significant amount of missing pixels and strong noise, limiting their usability in many computer vision applications. In this paper, we present an efficient hole filling and damaged region restoration method that improves the quality of the depth maps obtained with the Microsoft Kinect device. The proposed approach is based on a modified exemplar-based inpainting and LPA-ICI filtering by exploiting the correlation between color and depth values in local image neighborhoods. As a result, edges of the objects are sharpened and aligned with the objects in the color image. Several examples considered in this paper show the effectiveness of the proposed approach for large holes removal as well as recovery of small regions on several test images of depth maps. We perform a comparative study and show that statistically, the proposed algorithm delivers superior quality results compared to existing algorithms.

Keywords

  • Depth map, Filtering, Image processing, Inpainting, Kinect, Occlusion

Publication forum classification

Field of science, Statistics Finland