Depth guided image inpainting algorithm for free-viewpoint video

L. Ma, Q.L. Do, P.H.N. With, de

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

16 Citations (Scopus)
307 Downloads (Pure)

Abstract

Free-Viewpoint Video (FVV) is a novel technique which creates virtual images of multiple direction by view synthesis. In this paper, an exemplar-based depth-guided inpainting algorithm is proposed to fill disocclusions due to uncovered areas after projection. We develop an improved priority function which uses the depth information to impose a desirable inpainting order. We also propose an efficient background-foreground separation technique to enhance the accuracy of hole filling. Furthermore, a gradient-based searching approach is developed to reduce the computational cost and the location distance is incorporated into patch matching criteria to improve the accuracy. The experimental results have shown that the gradientbased search in our algorithm requires a much lower computational cost (factor of 6 compared to global search), while producing significantly improved visual results.
Original languageEnglish
Title of host publicationIEEE International Conference on Image Processing (ICIP), 30 September- 3 October 2012, Florida, USA
Place of PublicationOrlando, Florida, USA
PublisherInstitute of Electrical and Electronics Engineers
Pages1721-1724
ISBN (Print)978-1-4673-2532-5
Publication statusPublished - 2012
Event19th IEEE International Conference on Image Processing (ICIP 2012) - Lake Buena Vista, FL, United States
Duration: 30 Sept 20123 Oct 2012
Conference number: 19

Conference

Conference19th IEEE International Conference on Image Processing (ICIP 2012)
Abbreviated titleICIP 2012
Country/TerritoryUnited States
CityLake Buena Vista, FL
Period30/09/123/10/12

Fingerprint

Dive into the research topics of 'Depth guided image inpainting algorithm for free-viewpoint video'. Together they form a unique fingerprint.

Cite this