Part-Based Modelling of Compound Scenes from Images

Anton van den Hengel, Chris Russell, Anthony Dick, John Bastian, Daniel Pooley, Lachlan Fleming, Lourdes Agapito

CVPR 2015

doi:10.1109/CVPR.2015.7298689 /cvpr/2015/vandenhengel2015cvpr-partbased/

Abstract

We propose a method to recover the structure of a compound scene from multiple silhouettes. Structure is expressed as a collection of 3D primitives chosen from a pre-defined library, each with an associated pose. This has several advantages over a volume or mesh representation both for estimation and the utility of the recovered model. The main challenge in recovering such a model is the combinatorial number of possible arrangements of parts. We address this issue by exploiting the intrinsic structure and sparsity of the problem, and show that our method scales to scenes constructed from large libraries of parts.

PDF CVPR Semantic Scholar

Cite

Text

van den Hengel et al. "Part-Based Modelling of Compound Scenes from Images." Conference on Computer Vision and Pattern Recognition, 2015. doi:10.1109/CVPR.2015.7298689

Markdown

[van den Hengel et al. "Part-Based Modelling of Compound Scenes from Images." Conference on Computer Vision and Pattern Recognition, 2015.](https://mlanthology.org/cvpr/2015/vandenhengel2015cvpr-partbased/) doi:10.1109/CVPR.2015.7298689

BibTeX

@inproceedings{vandenhengel2015cvpr-partbased,
  title     = {{Part-Based Modelling of Compound Scenes from Images}},
  author    = {van den Hengel, Anton and Russell, Chris and Dick, Anthony and Bastian, John and Pooley, Daniel and Fleming, Lachlan and Agapito, Lourdes},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2015},
  doi       = {10.1109/CVPR.2015.7298689},
  url       = {https://mlanthology.org/cvpr/2015/vandenhengel2015cvpr-partbased/}
}