Estimating Building Age from Google Street View Images Using Deep Learning (Short Paper)

Authors Yan Li, Yiqun Chen, Abbas Rajabifard, Kourosh Khoshelham, Mitko Aleksandrov



PDF
Thumbnail PDF

File

LIPIcs.GISCIENCE.2018.40.pdf
  • Filesize: 424 kB
  • 7 pages

Document Identifiers

Author Details

Yan Li
  • Department of Infrastructure Engineering, University of Melbourne/Melbourne, Australia
Yiqun Chen
  • Department of Infrastructure Engineering, University of Melbourne/Melbourne, Australia
Abbas Rajabifard
  • Department of Infrastructure Engineering, University of Melbourne/Melbourne, Australia
Kourosh Khoshelham
  • Department of Infrastructure Engineering, University of Melbourne/Melbourne, Australia
Mitko Aleksandrov
  • Department of Infrastructure Engineering, University of Melbourne/Melbourne, Australia

Cite AsGet BibTex

Yan Li, Yiqun Chen, Abbas Rajabifard, Kourosh Khoshelham, and Mitko Aleksandrov. Estimating Building Age from Google Street View Images Using Deep Learning (Short Paper). In 10th International Conference on Geographic Information Science (GIScience 2018). Leibniz International Proceedings in Informatics (LIPIcs), Volume 114, pp. 40:1-40:7, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2018)
https://doi.org/10.4230/LIPIcs.GISCIENCE.2018.40

Abstract

Building databases are a fundamental component of urban analysis. However such databases usually lack detailed attributes such as building age. With a large volume of building images being accessible online via API (such as Google Street View), as well as the fast development of image processing techniques such as deep learning, it becomes feasible to extract information from images to enrich building databases. This paper proposes a novel method to estimate building age based on the convolutional neural network for image features extraction and support vector machine for construction year regression. The contributions of this paper are two-fold: First, to our knowledge, this is the first attempt for estimating building age from images by using deep learning techniques. It provides new insight for planners to apply image processing and deep learning techniques for building database enrichment. Second, an image-base building age estimation framework is proposed which doesn't require information on building height, floor area, construction materials and therefore makes the analysis process simpler and more efficient.

Subject Classification

ACM Subject Classification
  • Computing methodologies → Supervised learning by regression
Keywords
  • Building database
  • deep learning
  • CNN
  • SVM
  • Google Street View

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads

References

  1. Filip Biljecki and Maximilian Sindram. Estimating building age with 3d gis. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, IV-4/W5, 2017. Google Scholar
  2. Ali Borji, Ming-Ming Cheng, Huaizu Jiang, and Jia Li. Salient object detection: A benchmark. IEEE Transactions on Image Processing, 24(12):5706-5722, 2015. Google Scholar
  3. J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei. ImageNet: A Large-Scale Hierarchical Image Database. In CVPR09, 2009. Google Scholar
  4. Harris Drucker, Christopher JC Burges, Linda Kaufman, Alex J Smola, and Vladimir Vapnik. Support vector regression machines. In Advances in neural information processing systems, pages 155-161, 1997. Google Scholar
  5. Abhimanyu Dubey, Nikhil Naik, Devi Parikh, Ramesh Raskar, and César A Hidalgo. Deep learning the city: Quantifying urban perception at a global scale. In European Conference on Computer Vision, pages 196-212. Springer, 2016. Google Scholar
  6. Timnit Gebru, Jonathan Krause, Yilun Wang, Duyun Chen, Jia Deng, Erez Lieberman Aiden, and Li Fei-Fei. Using deep learning and google street view to estimate the demographic makeup of neighborhoods across the united states. Proceedings of the National Academy of Sciences, page 201700035, 2017. Google Scholar
  7. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770-778, 2016. Google Scholar
  8. Gao Huang, Zhuang Liu, Laurens van der Maaten, and Kilian Q Weinberger. Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017. Google Scholar
  9. Alex Krizhevsky and Geoffrey Hinton. Learning multiple layers of features from tiny images. Technical report, University of Toronto, 2009. Google Scholar
  10. Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems, pages 1097-1105, 2012. Google Scholar
  11. Yan Li, Majid Sarvi, Kourosh Khoshelham, and Milad Haghani. Real-time level-of-service maps generation from cctv videos. In Transportation Research Board 97th Annual Meeting, 2018. URL: https://trid.trb.org/view/1497380.
  12. Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Alban Desmaison, Luca Antiga, and Adam Lerer. Automatic differentiation in pytorch, 2017. URL: https://openreview.net/forum?id=BJJsrmfCZ.
  13. F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825-2830, 2011. Google Scholar
  14. Abbas Rajabifard. Spatial data infrastructure. International Federation of Surveyors (FIG), 2012. Google Scholar
  15. Jürgen Schmidhuber. Deep learning in neural networks: An overview. Neural networks, 61:85-117, 2015. Google Scholar
  16. Bolei Zhou, Agata Lapedriza, Aditya Khosla, Aude Oliva, and Antonio Torralba. Places: A 10 million image database for scene recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017. Google Scholar
Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail