Details
| Original language | English |
|---|---|
| Title of host publication | Computer Vision - ACCV 201 |
| Subtitle of host publication | 11th Asian Conference on Computer Vision, Revised Selected Papers |
| Pages | 611-623 |
| Number of pages | 13 |
| ISBN (electronic) | 978-3-642-37431-9 |
| Publication status | Published - 2013 |
| Event | 11th Asian Conference on Computer Vision, ACCV 2012 - Daejeon, Korea, Republic of Duration: 5 Nov 2012 → 9 Nov 2012 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Number | PART 3 |
| Volume | 7726 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (electronic) | 1611-3349 |
Abstract
Visual effect creation as used in movie production often require structure and motion recovery and video segmentation. Both techniques are essential to integrate virtual objects between scene elements. In this paper, a new method for video segmentation is presented. It incorporates 3D scene information from the structure and motion recovery. By connecting and evaluating discontinued feature tracks, occlusion and reappearance information is obtained during sequential camera and scene estimation. The foreground is characterized as image regions which temporarily occlude the rigid scene structure. The scene structure is represented by reconstructed object points. Their projections onto the camera images provide the cues for regions classified as foreground or background. The knowledge of occluded parts of a connected feature track is used to feed the object segmentation which crops the foreground image regions automatically. Two applications are presented: the occlusion of integrated virtual objects and the blurred background effect. Several demonstrations on official and self-made data show very realistic results in augmented reality.
ASJC Scopus subject areas
- Mathematics(all)
- Theoretical Computer Science
- Computer Science(all)
- General Computer Science
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
Computer Vision - ACCV 201: 11th Asian Conference on Computer Vision, Revised Selected Papers. 2013. p. 611-623 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7726 LNCS, No. PART 3).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Learning Object Appearance from Occlusions Using Structure and Motion Recovery
AU - Cordes, Kai
AU - Scheuermann, Björn
AU - Rosenhahn, Bodo
AU - Ostermann, Jörn
PY - 2013
Y1 - 2013
N2 - Visual effect creation as used in movie production often require structure and motion recovery and video segmentation. Both techniques are essential to integrate virtual objects between scene elements. In this paper, a new method for video segmentation is presented. It incorporates 3D scene information from the structure and motion recovery. By connecting and evaluating discontinued feature tracks, occlusion and reappearance information is obtained during sequential camera and scene estimation. The foreground is characterized as image regions which temporarily occlude the rigid scene structure. The scene structure is represented by reconstructed object points. Their projections onto the camera images provide the cues for regions classified as foreground or background. The knowledge of occluded parts of a connected feature track is used to feed the object segmentation which crops the foreground image regions automatically. Two applications are presented: the occlusion of integrated virtual objects and the blurred background effect. Several demonstrations on official and self-made data show very realistic results in augmented reality.
AB - Visual effect creation as used in movie production often require structure and motion recovery and video segmentation. Both techniques are essential to integrate virtual objects between scene elements. In this paper, a new method for video segmentation is presented. It incorporates 3D scene information from the structure and motion recovery. By connecting and evaluating discontinued feature tracks, occlusion and reappearance information is obtained during sequential camera and scene estimation. The foreground is characterized as image regions which temporarily occlude the rigid scene structure. The scene structure is represented by reconstructed object points. Their projections onto the camera images provide the cues for regions classified as foreground or background. The knowledge of occluded parts of a connected feature track is used to feed the object segmentation which crops the foreground image regions automatically. Two applications are presented: the occlusion of integrated virtual objects and the blurred background effect. Several demonstrations on official and self-made data show very realistic results in augmented reality.
UR - http://www.scopus.com/inward/record.url?scp=84875879880&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-37431-9_47
DO - 10.1007/978-3-642-37431-9_47
M3 - Conference contribution
AN - SCOPUS:84875879880
SN - 9783642374302
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 611
EP - 623
BT - Computer Vision - ACCV 201
T2 - 11th Asian Conference on Computer Vision, ACCV 2012
Y2 - 5 November 2012 through 9 November 2012
ER -