Dense Structure-from-Motion: An Approach Based on Segment Matching

Abstract

For 3-D video applications, dense depth maps are required. We present a segment-based structure-from-motion technique. After image segmentation, we estimate the motion of each segment. With knowledge of the camera motion, this can be translated into depth. The optimal depth is found by minimizing a suitable error norm, which can handle occlusions as well. This method combines the advantages of motion estimation on the one hand, and structure-from-motion algorithms on the other hand. The resulting depth maps are pixel-accurate due to the segmentation, and have a high accuracy: depth differences corresponding to motion differences of 1/8^th of a pixel can be recovered.

Cite

Text

Ernst et al. "Dense Structure-from-Motion: An Approach Based on Segment Matching." European Conference on Computer Vision, 2002. doi:10.1007/3-540-47967-8_15

Markdown

[Ernst et al. "Dense Structure-from-Motion: An Approach Based on Segment Matching." European Conference on Computer Vision, 2002.](https://mlanthology.org/eccv/2002/ernst2002eccv-dense/) doi:10.1007/3-540-47967-8_15

BibTeX

@inproceedings{ernst2002eccv-dense,
  title     = {{Dense Structure-from-Motion: An Approach Based on Segment Matching}},
  author    = {Ernst, Fabian and Wilinski, Piotr and van Overveld, Cornelius W. A. M.},
  booktitle = {European Conference on Computer Vision},
  year      = {2002},
  pages     = {217-231},
  doi       = {10.1007/3-540-47967-8_15},
  url       = {https://mlanthology.org/eccv/2002/ernst2002eccv-dense/}
}