Temporal Surface Reconstruction
Abstract
An approach which allows an arbitrary structure estimation to be embedded into a recursive estimation process that incrementally improves a structure estimate with every new frame that becomes available is discussed. The approach is based on Bayesian estimation theory and the Kalman filter. The authors demonstrate how it may be applied to such domains as depth from motion and depth from shading.<<ETX>>
Cite
Text
Heel. "Temporal Surface Reconstruction." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1991. doi:10.1109/CVPR.1991.139761Markdown
[Heel. "Temporal Surface Reconstruction." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1991.](https://mlanthology.org/cvpr/1991/heel1991cvpr-temporal/) doi:10.1109/CVPR.1991.139761BibTeX
@inproceedings{heel1991cvpr-temporal,
title = {{Temporal Surface Reconstruction}},
author = {Heel, Joachim},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {1991},
pages = {607-612},
doi = {10.1109/CVPR.1991.139761},
url = {https://mlanthology.org/cvpr/1991/heel1991cvpr-temporal/}
}