The ICA Commissions on Visual Analytics and Cognitive Issues in Geographic Information Visualization are pleased to announce a Workshop on Cartography and AI (MapAI 2023) taking place on August 12, 2023 at Stellenbosch University, South Africa before the ICC 2023 in Cape Town, South Africa.
Amy L. Griffin, RMIT
Anthony C. Robinson, Penn State University
Arzu Çöltekin, FHNW – University of Applied Sciences Northwestern Switzerland
Cartographers have assessed the potential use of artificial intelligence for mapping for decades. Early work on expert systems explored AI as early as the 1980s (e.g., Buttenfield 1984; Fisher & Mackaness 1987; Brassel & Weibel, 1988; Johnson & Basoglu 1989). The AI tools of the time were limited by the (lack of) availability of computing power and data. More recently, as AI tools have become both more powerful and easier to use, a few cartographers and computer scientists have begun experimenting with artificial intelligence technologies to see how they might be applied to maps and mapping processes (e.g., Kang et al., 2019; Zhao et al, 2021; Christophe et al., 2022; Zhou et al, 2022; Santos et al., 2023). Others have made initial efforts to review the potential of AI technologies for cartography (Kang et al., 2022), laying out some possibilities and also some points of caution by identifying ethical issues these technologies raise and/or exacerbate. The 2022 launches of DALL·E 2, ChatGPT, and other AI platforms have caught the attention of the general public by making artificial intelligence technologies easy to use for a range of everyday tasks. Some cartographers have already put these to use for assisting their mapping practice (see, for example, https://www.youtube.com/watch?v=OCOpxy3wk-o). There is much more for cartographers to explore, including the potential impacts of the use of AI on map users’ and map makers’ cognitive processes (see Keskin & Kettunen, 2023 for an initial investigation).
In this workshop, we aim to bring together ideation and practical experimentation to collaboratively explore some of the potential and limits of current AI technologies for cartographic practice and map use.
Call for Presentations
The first half of our planned 1-day workshop will be for participants to present Lightning Talks. In 5 minutes presenters will showcase either one major challenge or one significant opportunity you see that intersects between AI and Cartography.
These presentations should focus on frontiers in cartographic research that intersect with AI tools or techniques, and creative/unorthodox approaches are welcomed. Work-in-progress is the intended target, versus projects that are already fully completed.
Example topics could include, but are not limited to:
- Implications of deepfake maps and satellite images
- Machine-learning / AI based map updating based on image input
- AI-generated wayfinding directions
- Geographic aspects of algorithmic bias
- Automated means of iterating design elements in cartographic layouts
- Explainable AI & mapping
- Natural language interaction with maps
- Mapping with ChatGPT, DALL-E, Midjourney, Stable Diffusion, etc…
- Human-machine collaboration using maps
Workshop presentations will be used to motivate group discussions and hands-on experimentation in the second half of the planned 1-day workshop. We are excited to work together to push the limits of various AI mechanisms for cartographic design and inquiry, learning along the way about where the frontiers lie for future research and applications of AI in Cartography.
Deadline Extended! Please submit an abstract of <250 words that showcases either one major challenge or one significant opportunity that intersects AI and Cartography by
May 15, 2023 May 19, 2023 to https://easychair.org/conferences/?conf=mapai2023. All submissions will be reviewed by the workshop organizers for clarity and fit with workshop themes. A final workshop agenda including accepted talks will be communicated by June 23, 2023.
This workshop will take place in-person on August 12, 2023 at the Department of Geography and Environmental Studies, Stellenbosch University, Stellenbosch, South Africa. If you are attending the ICC 2023 in Cape Town, Stellenbosch is roughly an hour away by Taxi/Uber/Shuttle Service. We recommend staying in Stellenbosch for 1 or 2 nights if you prefer not to commute to/from Cape Town. Stellenbosch is famous for its wineries and there are many scenic hotels located on wineries nearby.
The workshop will be held at the Narga D Computer Lab at the Chamber of Mines (Geology) Building at Stellenbosch University, located here. This workshop is planned to take place in-person. We are exploring options to provide an online stream / recording of some of the talks, but the primary means for engagement will be to work together on-site.
We will provide an update closer to the start of the workshop with details regarding support for meals/coffee breaks on-site. The organizers will do everything possible to support food and drink while minimizing participation fees for participants.
We will notify authors and publish a preliminary workshop schedule by June 23, 2023.
Getting to Stellenbosch
We recommend booking a shuttle service or taking an Uber to get to Stellenbosch from downtown Cape Town or from Cape Town International Airport. Shuttle service and Uber costs are comparable, and should be roughly 600 ZAR per person between Cape Town Airport and Stellenbosch (~30 USD) and between 900 ZAR per person between Cape Town city center and Stellenbosch (~50 USD).
Stellenbosch Shuttles: https://www.stellenboschshuttles.co.za/book-now
Shuttle Up: https://shuttleup.co.za/
Uber price estimates + pre-booking: https://www.uber.com/global/en/price-estimate/
Staying in Stellenbosch
This is a volunteer-organized pre-conference workshop, so we do not have resources to support a room block at a single hotel. Instead, we recommend you consider the following options:
Downtown, within walking distance (1km) from workshop venue:
Oude Werf – 4*, rooms available for ~$125 USD per night
Stellenbosch Hotel – 3*, rooms available for ~$75 USD per night
Wine Estates, within 10min Taxi/Uber from workshop venue (~7km to the NE):
Clouds Estate – 5*, rooms available for ~$175 USD per night
Alluvia Boutique Winery – 4*, rooms available for ~$125 per night
Le Pommier Wines Estate – 4*, rooms available for ~$100 per night.
Buttenfield, B. P. (1984). Line structure in graphic and geographic space (computer cartography, artificial intelligence, generalization). University of Washington. PhD Dissertation.
Brassel, K. E., & Weibel, R. (1988). A review and conceptual framework of automated map generalization. International Journal of Geographical Information Systems, 2(3), 229-244.
Christophe, S., Mermet, S., Laurent, M., & Touya, G. (2022). Neural map style transfer exploration with GANs. International Journal of Cartography, 8(1), 18-36.
Fisher, P. F., & Mackaness, W. A. (1987). Are cartographic expert systems possible?. In Proceedings AutoCarto Vol. 8, pp. 530-534.
Johnson, D. S., & Basoglu, U. (1989). The use of artificial intelligence in the automated placement of cartographic names. In Proceedings Auto-Carto Vol. 9, 225-230.
Kang, Y., Gao, S., & Roth, R. E. (2019). Transferring multiscale map styles using generative adversarial networks. International Journal of Cartography, 5(2–3), 115–141. https://doi.org/10.1080/23729333.2019.1615729
Kang, Y., Gao, S., & Roth, R. A Review and Synthesis of Recent GeoAI Research for Cartography: Methods, Applications, and Ethics. Proceedings of AutoCarto 2022. November 2-4, Redlands, CA.
Keskin, M., & Kettunen, P. (2023). Potential of eye-tracking for interactive geovisual exploration aided by machine learning. International Journal of Cartography, 1-23.
Santos, A., Martins, T., Correia, J. (2023). Creating stylised maps with neural style transfer. https://cdv.dei.uc.pt/stylised-maps/. Accessed 19 February 2023.
Zhao, B., Zhang, S., Xu, C., Sun, Y., & Deng, C. (2021). Deep fake geography? When geospatial data encounter Artificial Intelligence. Cartography and Geographic Information Science, 48(4), 338-352.
Zhou, Z., Fu, C., & Weibel, R. (2022). Building simplification of vector maps using graph convolutional neural networks. Abstracts of the ICA, 5, 1-2.