Learn More
We propose a multi-view photometric stereo technique that uses photometric normal consistency to jointly estimate surface position and orientation. The underlying scene representation is based on oriented points, yielding more flexibility compared to smoothly varying surfaces. We demonstrate that the often employed least squares error of the Lambertian(More)
This is a slightly updated version of a conference submission that was rejected in 2008. Since on one hand GPU technology is progressing very rapidly and on the other hand this paper is referenced on the Middlebury Multi-view Stereo benchmark page (http:// vision.middlebury.edu/mview/) we decided to finally publish it as a technical report. Abstract In(More)
The ultimate goal of many image-based modeling systems is to render photo-realistic novel views of a scene without visible artifacts. Existing evaluation metrics and benchmarks focus mainly on the geometric accuracy of the reconstructed model, which is, however, a poor predictor of visual accuracy. Furthermore, using only geometric accuracy by itself does(More)
  • 1