Supplementary materials for the publication "Generative text-to-image diffusion for map creation based on geosocial media data."
datacite.FundingReference.funderName | Deutsche Forschungsgemeinschaft | |
Contributing person | Institut für Kartographie | |
Contributing person | Burghardt, Dirk (orcid: 0000-0003-2949-4887) | |
datacite.description.TableOfContents | The repository contains: - Input Data: Shapefiles, Word2Vec Language Model, Intermediate files - Code: *.py - Jupyter Notebooks: *.ipynb (Code) - Jupyter Notebooks: *.md (code, comments) - Jupyter Notebooks: *.html (code, comments, output, for archive purposes) - Output: png (figures and graphics) - Output: svg (selected figures and graphics) - Chore: Other files necessary to reproduce figures and output | |
Documentation of the data | Methods: The workflow to load and process the data provided is available in Jupyter Notebooks. Data Acquisition: Data query from public Application Programming Interfaces (APIs). Data Processing: We demonstrate an integrated workflow that uses text-to-image Stable Diffusion as its core to automatically produce icon maps. The provided workflow is based on aggregating geosocial media data from Twitter, Flickr, Instagram, and iNaturalist. This data is used to account for the population's collective attribution of meaning and importance in map generation. The complete code to reproduce the workflow is shared in several Jupyter notebooks herein. | |
Additional geographical or spatial references | 51.038885, 13.761159 | |
Additional geographical or spatial references | Großer Garten Dresden, Campus TU Dresden | |
Additional geographical or spatial references | Sachsen | |
Countries to which the data refer | GERMANY | |
References to related material | Publication "Generative text-to-image diffusion for map creation based on geosocial media data." (Kartographische Nachrichten) | |
Description of the data | The repository contains: - Input Data: Shapefiles, Word2Vec Language Model, Intermediate files - Code: *.py - Jupyter Notebooks: *.ipynb (Code) - Jupyter Notebooks: *.md (code, comments) - Jupyter Notebooks: *.html (code, comments, output, for archive purposes) - Output: png (figures and graphics) - Output: svg (selected figures and graphics) - Chore: Other files necessary to reproduce figures and output The base data is from Stability AI, Stable Diffusion 1.5, which was trained on LAION-5B (Schuhmann et al. 2022), a dataset consisting of 5.85 billion images scraped from the web. | |
Type of the data | Text | |
Type of the data | Software | |
Type of the data | Image | |
Type of the data | Model | |
Type of the data | Dataset | |
Type of the data | Other | |
Total size of the dataset | 622035578 | |
Author | Dunkel, Alexander | |
Author | Gugulica, Madalina | |
Upload date | 2023-11-03T12:38:31Z | |
Publication date | 2023-11-03T12:38:31Z | |
Publication date | 2026-06-12T12:44:03Z | |
Data of data creation | 2023 | |
Publication date | 2023-11-03 | |
Abstract of the dataset | Supporting Information for the publication "Generative text-to-image diffusion for map creation based on geosocial media data"(Release v1.0.0). Updates from the peer review process may not be reflected in this submission. The original git repository with the latest versions can be found at https://gitlab.hrz.tu-chemnitz.de/ad/mapnik_stablediffusion | |
Public reference to this page | https://opara.zih.tu-dresden.de/handle/123456789/2707 | |
Public reference to this page | https://doi.org/10.25532/OPARA-253 | |
dc.language | eng | |
Publisher | Technische Universität Dresden | |
Licence | Attribution 4.0 International | |
URI of the licence text | http://creativecommons.org/licenses/by/4.0/ | |
Specification of the discipline(s) | 3::34::317 | |
Specification of the discipline(s) | 3::34 | |
Specification of the discipline(s) | 4::44::409 | |
Title of the dataset | Supplementary materials for the publication "Generative text-to-image diffusion for map creation based on geosocial media data." | |
dc.title.alternative | Release v1.0.0 | |
Research instruments | Carto-Lab Docker v0.15.7 (https://gitlab.vgiscience.de/lbsn/tools/jupyterlab) | |
Research instruments | TUD ZIH HPC Cluster, alpha-interactive | |
Underlying research object | Aggregated, clustered, privacy-friendly geosocial media data | |
Software | Stable-Diffusion (Version 1.5) | |
Project abstract | "Geovisual analysis of VGI for understanding people’s behaviour in relation to multi-faceted context" Volunteered Geographic Information (VGI) in the form of actively and passively generated spatial content offers extensive potential for a wide range of applications. Realising this potential however requires methods which take account of the specific properties of such data, for example its heterogeneity, quality, subjectivity, spatial resolution and temporal relevance. The creation and production of such content through social media platforms is an expression of human behaviour, and as such influenced strongly by events and external context. In this project we will develop geovisual analysis methods which show how actors interact in LBSM, and how their interactions influence, and are influenced by, their physical and social environment and relations. | |
Project title | EvaVGI (EvaVGI) |
Files
Original bundle
Collections
