Structured Outdoor Architecture Reconstruction by Exploration and Classification
Abstract
This paper presents an explore-and-classify framework for structured architectural reconstruction from aerial image. Starting from a potentially imperfect building reconstruction by an existing algorithm, our approach 1) explores the space of building models by modifying the reconstruction via heuristic actions; 2) learns to classify the correctness of building models while generating classification labels based on the ground-truth; and 3) repeat. At test time, we iterate exploration and classification, seeking for a result with the best classification score. We evaluate the approach using initial reconstructions by two baselines and two state-of-the-art reconstruction algorithms. Qualitative and quantitative evaluations demonstrate that our approach consistently improves the reconstruction quality from every initial reconstruction.
Cite
Text
Zhang et al. "Structured Outdoor Architecture Reconstruction by Exploration and Classification." International Conference on Computer Vision, 2021. doi:10.1109/ICCV48922.2021.01220Markdown
[Zhang et al. "Structured Outdoor Architecture Reconstruction by Exploration and Classification." International Conference on Computer Vision, 2021.](https://mlanthology.org/iccv/2021/zhang2021iccv-structured/) doi:10.1109/ICCV48922.2021.01220BibTeX
@inproceedings{zhang2021iccv-structured,
title = {{Structured Outdoor Architecture Reconstruction by Exploration and Classification}},
author = {Zhang, Fuyang and Xu, Xiang and Nauata, Nelson and Furukawa, Yasutaka},
booktitle = {International Conference on Computer Vision},
year = {2021},
pages = {12427-12435},
doi = {10.1109/ICCV48922.2021.01220},
url = {https://mlanthology.org/iccv/2021/zhang2021iccv-structured/}
}