Data and code underlying the PhD thesis: Pixels · People · Places: Computer Vision and Image Embeddings for Perception-Aware Urban Analytics

This repository contains:

Scripts for collecting different layers of urban data systematically (Land Uses, Point of Interests, Street Networks, Panoids (Google Street View)Scripts for performing different urban and Image analysis: Object Detection, Spatial Aggregation, Regressions, (spatial) Correlations, Embeddings generation, among others.Post-processed data required for the analysis conducted in the studiesFor more info, check out the README file in the repository.

The codes were used to conduct different studies at Delft University of Technology, as part of Francisco Garrido-Valenzuela's PhD Thesis project (2021-2025).