StudiumOffene Abschlussarbeiten
Estimating House Prices from Multiple Data Sources

Estimating House Prices from Multiple Data Sources

Leitung:  Feng, Sester
Jahr:  2020

Goal of this thesis

Google Streetview and satellite imagery are nowadays well available for many cities. These data contain the information about buildings, including age, material, and condition. With these visual features, Law et al. (2019) achieved a better house price estimation compared to only using the traditional housing features. Meanwhile, Gebru et al. (2017) discovered the strong association between socioeconomic attributes and cars detected by deep learning algorithm in Google Streetview images.

The aim of this work is to estimate house prices of a region with multiple data sources, namely frontage Streetview images, building and cars detected in Streetview images, satellite imagery, and socio-economic data. A data fusion process is to be performed and the benefits of different data sources are to be analyzed.



1. Literature review and data investigation

2. Prepare Streetview and satellite images for a study area

3. Detection of objects with additional attribute from Streetview images (e.g. buildings, cars, and car brand)

4. Train model and evaluate different data sources

5. Prepare final thesis



► Pipeline to download and pre-process Google Streetview data

► OSM data, public house price data

► Related literature



► Programming skills (preferably Python)

► Knowledge and experience in deep learning (preferably pytorch)

► Knowledge on GIS and geospatial data processing (e.g. QGIS, geopandas, shapely)



► Gebru, T., Krause, J., Wang, Y., Chen, D., Deng, J., Aiden, E. L., & Fei-Fei, L. (2017). 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, 114(50), 13108-13113.

► Law, S., Paige, B., & Russell, C. (2019). Take a look around: using street view and satellite images to estimate house prices. ACM Transactions on Intelligent Systems and Technology (TIST), 10(5), 1-19.



► Yu Feng (, 0511 762-19437)

► Prof. Dr.-Ing. habil. Monika Sester (, 0511 762-3588)