ML Anthology

A Lighting-Invariant Point Processor for Shading

Kathryn Heal, Jialiang Wang, Steven J. Gortler, Todd Zickler
CVPR 2020
doi:10.1109/CVPR42600.2020.00017 /cvpr/2020/heal2020cvpr-lightinginvariant/

Abstract

Under the conventional diffuse shading model with unknown directional lighting, the set of quadratic surface shapes that are consistent with the spatial derivatives of intensity at a single image point is a two-dimensional algebraic variety embedded in the five-dimensional space of quadratic shapes. We describe the geometry of this variety, and we introduce a concise feedforward model that computes an explicit, differentiable approximation of the variety from the intensity and its derivatives at any single image point. The result is a parallelizable processor that operates at each image point and produces a lighting-invariant descriptor of the continuous set of compatible surface shapes at the point. We describe two applications of this processor: two-shot uncalibrated photometric stereo and quadratic-surface shape from shading.

PDF CVPR Semantic Scholar

Cite

Text

Heal et al. "A Lighting-Invariant Point Processor for Shading." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020. doi:10.1109/CVPR42600.2020.00017

Markdown

[Heal et al. "A Lighting-Invariant Point Processor for Shading." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020.](https://mlanthology.org/cvpr/2020/heal2020cvpr-lightinginvariant/) doi:10.1109/CVPR42600.2020.00017

BibTeX

@inproceedings{heal2020cvpr-lightinginvariant,
  title     = {{A Lighting-Invariant Point Processor for Shading}},
  author    = {Heal, Kathryn and Wang, Jialiang and Gortler, Steven J. and Zickler, Todd},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2020},
  doi       = {10.1109/CVPR42600.2020.00017},
  url       = {https://mlanthology.org/cvpr/2020/heal2020cvpr-lightinginvariant/}
}