ABSTRACT SUBMISSION DEADLINE EXTENDED:
10th January, 18:00 GMT 24th January, 18:00 GMT
SpaceTimeAI is being organised in partnership with the 25th Geographical Information Science Research UK Conference. Discounts are available for joint registration for both conferences.
In recent years, the cross-pollination of research between the fields of artificial intelligence, machine learning and GIScience has led to the emergence of a rich vein of research that has been termed by many as GeoAI. GeoAI involves the use of advanced AI algorithms, such as convolutional and recurrent neural networks, for solving complex geographical problems. These methods have shown excellent progress across many application domains, including satellite image classification, semantic segmentation, simultaneous localisation and mapping (SLAM) and object detection, amongst others. As we look towards the next decades, AI will underpin many of the functions that we will come to rely on, including autonomous vehicles, smart cities, weather and climate modelling and infrastructure networks. Crucial to the success of these approaches will be the ability to model the properties of spatio-temporal data – so called SpaceTimeAI. In particular, SpaceTimeAI has an important role to play in the simulation of human and environmental processes through the use of new multi-agent reinforcement learning (MARL) and generative adversarial network (GAN) methods.
This symposium aims to bring together the ISPRS and wider community working in SpaceTimeAI and GeoAI, geocomputation, geosimulation, spatial and spatio-temporal analysis and related disciplines to present their latest research. The event will provide a forum to discuss the issues surrounding the ubiquity of AI in all aspects of daily life. Extended abstracts are encouraged that address the following topics:
- AI algorithms for spatial and spatio-temporal data, including deep learning, computer vision, deep reinforcement learning and generative adversarial networks.
- Applications of AI in the geosciences, including remote sensing, BIM, smart cities, urban analytics, transport, environmental modelling and related topics.
- Geosimulation, including cellular automata, agent based modelling and multi-agent reinforcement learning.
- Geocomputation for spatio-temporal data
- Spatio-temporal data modelling and analysis
- Ethical Issues of AI
- Comparisons of AI with other methods for spatio-temporal data, such as statistical or machine learning algorithms
- Generalisation and validation
- Benchmark dataset generation
- Integration of AI with existing software, such as GIS
- Approaches to teaching AI in the GIScience curriculum
If you are unsure about the relevance of your paper, please do not hesitate to contact the organisers at .
SpaceTimeAI is accepting submissions of abstracts and full papers. As the conference topic is an emerging field, the submission of a full paper is not required for presentation at the conference, but only full papers will be published in the ISPRS Archives.
Abstracts of a maximum length of 1000 words must be submitted via Easychair before 18:00 on
Friday 10th January Friday 24th January 2020. Abstracts should contain enough detail to
clearly convey the approach and (preliminary) results of the research, concepts, and the reasons why it should be considered an interesting subject for the audience.
Full Paper SubmissionFor inclusion in the ISPRS Archives, a full paper must be submitted. Papers of a maximum length of 8 pages (including references) must be submitted through EasyChair before 18:00 on Friday 28th February 2020. Submissions should be produced in accordance with the ISPRS Archives style guidelines, as per the templates below. All submissions should be in pdf format.
All papers will be subject to peer review and allocated to oral and poster sessions accordingly.
Publication and Proceedings
The conference proceedings will be published in
- Submissions Open: 18th November 2019
- Abstract Submission Deadline:
10th January 2020, 18:00 GMT24th January 2020, 18:00 GMT
- Full Paper Submission Deadline: 28th February 2020, 18:00 GMT
- Notification of Acceptance: Mid-February 2020
- Main Conference: 20th to 22nd April 2020