Details
| Original language | English |
|---|---|
| Title of host publication | Trends and Topics in Computer Vision |
| Subtitle of host publication | ECCV 2010 Workshops, Revised Selected Papers |
| Pages | 41-49 |
| Number of pages | 9 |
| ISBN (electronic) | 978-3-642-35740-4 |
| Publication status | Published - 2012 |
| Event | 11th European Conference on Computer Vision, ECCV 2010 - Heraklion, Crete, Greece Duration: 10 Sept 2010 → 11 Sept 2010 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Number | PART 2 |
| Volume | 6554 |
| ISSN (Print) | 0302-9743 |
| ISSN (electronic) | 1611-3349 |
Abstract
When recording presentations which include visualizations displayed on a monitor or with a video projector, the quality of the captured video suffers from color distortion and aliasing effects in the display area. A photometric calibration for the whole image can not compensate for these defects. In this paper, we present a per-pixel photometric calibration method that solves this problem.We measure the joint monitor-camera response function for every single camera pixel by displaying red, green, and blue screens at all brightness levels and capture them separately. These measurements are used to estimate the joint response function for every single pixel and all three color channels with the empirical model of response (EMoR). We apply the estimated response functions on subsequent captures of the display to calibrate them.Our method achieves a mean absolute error of about 0.66 brightness levels, averaged over all pixels of the image. The performance is also demonstrated with a calibration of a real captured photo, which is hardly distinguishable from the original.
ASJC Scopus subject areas
- Mathematics(all)
- Theoretical Computer Science
- Computer Science(all)
- General Computer Science
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
Trends and Topics in Computer Vision : ECCV 2010 Workshops, Revised Selected Papers. 2012. p. 41-49 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6554 , No. PART 2).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Photometric Color Calibration of the Joint Monitor-Camera Response Function
AU - Elbrandt, Tobias
AU - Ostermann, Jörn
PY - 2012
Y1 - 2012
N2 - When recording presentations which include visualizations displayed on a monitor or with a video projector, the quality of the captured video suffers from color distortion and aliasing effects in the display area. A photometric calibration for the whole image can not compensate for these defects. In this paper, we present a per-pixel photometric calibration method that solves this problem.We measure the joint monitor-camera response function for every single camera pixel by displaying red, green, and blue screens at all brightness levels and capture them separately. These measurements are used to estimate the joint response function for every single pixel and all three color channels with the empirical model of response (EMoR). We apply the estimated response functions on subsequent captures of the display to calibrate them.Our method achieves a mean absolute error of about 0.66 brightness levels, averaged over all pixels of the image. The performance is also demonstrated with a calibration of a real captured photo, which is hardly distinguishable from the original.
AB - When recording presentations which include visualizations displayed on a monitor or with a video projector, the quality of the captured video suffers from color distortion and aliasing effects in the display area. A photometric calibration for the whole image can not compensate for these defects. In this paper, we present a per-pixel photometric calibration method that solves this problem.We measure the joint monitor-camera response function for every single camera pixel by displaying red, green, and blue screens at all brightness levels and capture them separately. These measurements are used to estimate the joint response function for every single pixel and all three color channels with the empirical model of response (EMoR). We apply the estimated response functions on subsequent captures of the display to calibrate them.Our method achieves a mean absolute error of about 0.66 brightness levels, averaged over all pixels of the image. The performance is also demonstrated with a calibration of a real captured photo, which is hardly distinguishable from the original.
UR - http://www.scopus.com/inward/record.url?scp=84871133213&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-35740-4_4
DO - 10.1007/978-3-642-35740-4_4
M3 - Conference contribution
AN - SCOPUS:84871133213
SN - 9783642357398
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 41
EP - 49
BT - Trends and Topics in Computer Vision
T2 - 11th European Conference on Computer Vision, ECCV 2010
Y2 - 10 September 2010 through 11 September 2010
ER -