ML Anthology

Bundle Pooling for Polygonal Architecture Segmentation Problem

Huayi Zeng, Kevin Joseph, Adam Vest, Yasutaka Furukawa
CVPR 2020
doi:10.1109/CVPR42600.2020.00083 /cvpr/2020/zeng2020cvpr-bundle/

Abstract

This paper introduces a polygonal architecture segmentation problem, proposes bundle-pooling modules for line structure reasoning, and demonstrates a virtual remodeling application that produces production quality results. Given a photograph of a house with a few vanishing point candidates, we decompose the house into a set of architectural components, each of which is represented as a simple geometric primitive. A bundle-pooling module pools convolutional features along a bundle of line segments (e.g., a family of vanishing lines) and fuses the bundle of features to determine polygonal boundaries or assign a corresponding vanishing point. Qualitative and quantitative evaluations demonstrate significant improvements over the existing techniques based on our metric and benchmark dataset. We will share the code and data for further research.

PDF CVPR Semantic Scholar

Cite

Text

Zeng et al. "Bundle Pooling for Polygonal Architecture Segmentation Problem." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020. doi:10.1109/CVPR42600.2020.00083

Markdown

[Zeng et al. "Bundle Pooling for Polygonal Architecture Segmentation Problem." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020.](https://mlanthology.org/cvpr/2020/zeng2020cvpr-bundle/) doi:10.1109/CVPR42600.2020.00083

BibTeX

@inproceedings{zeng2020cvpr-bundle,
  title     = {{Bundle Pooling for Polygonal Architecture Segmentation Problem}},
  author    = {Zeng, Huayi and Joseph, Kevin and Vest, Adam and Furukawa, Yasutaka},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2020},
  doi       = {10.1109/CVPR42600.2020.00083},
  url       = {https://mlanthology.org/cvpr/2020/zeng2020cvpr-bundle/}
}