Ctrl K

Code underlying the publication: From pixels to perceptions: using human similarity judgments to enrich urban space embeddings

Code underlying the publication: From pixels to perceptions: using human similarity judgments to enrich urban space embeddings

3
contributors

Description

This is the official repository for generating urban space embeddingsusingstreet-level imagery (SLI) and human similarity judgments. To use this framework, you must first conduct a similarity judgment experiment in which participants compare urban scenes. This repository provides all the tools needed to select the urban regions, design the experiment, implement the experiment, train our image embedding model and apply it on any city with the available data.

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).

Logo of Code underlying the publication: From pixels to perceptions: using human similarity judgments to enrich urban space embeddings
Keywords
Programming languages
  • Python 81%
  • Jupyter Notebook 7%
  • Markdown 7%
  • Other 4%
License
  • CC-BY-NC-SA-4.0
</>Source code
Packages
data.4tu.nl

Reference papers

Contributors

SvC
null Sander van Cranenburgh
OC
Oded Cats

Member of community

4TU