Beyond Feature Points: Structured Prediction for Monocular Non-Rigid 3D Reconstruction
Abstract
Existing approaches to non-rigid 3D reconstruction either are specifically designed for feature point correspondences, or require a good shape initialization to exploit more complex image likelihoods. In this paper, we formulate reconstruction as inference in a graphical model, where the variables encode the rotations and translations of the facets of a surface mesh. This lets us exploit complex likelihoods even in the absence of a good initialization. In contrast to existing approaches that set the weights of the likelihood terms manually, our formulation allows us to learn them from as few as a single training example. To improve efficiency, we combine our structured prediction formalism with a gradient-based scheme. Our experiments show that our approach yields tremendous improvement over state-of-the-art gradient-based methods.
Cite
Text
Salzmann and Urtasun. "Beyond Feature Points: Structured Prediction for Monocular Non-Rigid 3D Reconstruction." European Conference on Computer Vision, 2012. doi:10.1007/978-3-642-33765-9_18Markdown
[Salzmann and Urtasun. "Beyond Feature Points: Structured Prediction for Monocular Non-Rigid 3D Reconstruction." European Conference on Computer Vision, 2012.](https://mlanthology.org/eccv/2012/salzmann2012eccv-beyond/) doi:10.1007/978-3-642-33765-9_18BibTeX
@inproceedings{salzmann2012eccv-beyond,
title = {{Beyond Feature Points: Structured Prediction for Monocular Non-Rigid 3D Reconstruction}},
author = {Salzmann, Mathieu and Urtasun, Raquel},
booktitle = {European Conference on Computer Vision},
year = {2012},
pages = {245-259},
doi = {10.1007/978-3-642-33765-9_18},
url = {https://mlanthology.org/eccv/2012/salzmann2012eccv-beyond/}
}