GeoAI Challenges: Vegetation Mapping and Human Settlement Detection Challenge
* Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay
The National Institute of Statistics and Geography (INEGI), the International Telecommunication Union (ITU), AI for Good, the United Nations Global Geospatial Information Management (UN-GGIM) Academic Network and UN Geospatial Network are thrilled to announce the launch of two exciting GeoAI Challenges. The winners of these challenges will be awarded at the prestigious UN-GGIM Seventh High-Level Forum, to be held in Mexico City, on October 8-10, 2024. We invite students and professionals from around the world, including mapping agencies experts to participate and tackle real-world issues using geospatial technologies/science, AI and machine learning to accelerate progress on the implementation and monitoring of the United Nations Sustainable Development Goals (SDGs).
Don’t miss our upcoming webinar, where we will unveil the two challenge problem statements!
Human Settlement Detection Challenge
Human settlements are rapidly expanding, creating an urgent need for efficient and accurate monitoring systems to track and manage this growth The objective of this challenge is to develop a robust and accurate machine learning model that can detect human settlements in satellite imagery. The task is to build a binary classification model that can predict the presence of human settlements in the given test data. The evaluation of these predictions will determine the winner of the challenge. Participants are expected to submit their final model along with its predictions on the test dataset.
Vegetation Mapping Challenge
Vegetation is essential to biodiversity, carbon capture and storage, soil formation and protection and the regulation of the water cycle The objective of this challenge is to propose and develop a robust and accurate machine learning model that can help to clean and improve training data, either identifying outliers (wrong or suspicious labels), or even suggesting a more plausible label in the given test data. The evaluation of these predictions will determine the winner of the challenge. Participants are expected to submit their final model along with its predictions on the test data indicating outlier observations and desirably a suggested label.
For both challenges, ITU provides a state-of-the-art, free-of-charge computing platform to participants of the Challenge who do not have adequate access to computing platform and resources in their respective institutions. The computing platform will provide participants access to:
- Free GPUs and CPUs
- Hosted Jupyter Notebook server
- Python kernel
- Pre-installed machine learning packages, e.g. PyTorch and TensorFlow
Seventh High-Level Forum of UN-GGIM
Geospatial information is increasingly recognized as a critical component of national, regional, and global information infrastructures. Geospatial data serves as a key element of national infrastructure and the knowledge economy, providing a model of what is happening and where it is occurring. It also serves to integrate a wide range of government services that contribute to national prosperity, security, and sustainable economic, social, and environmental development.
In July 2011, the United Nations Economic and Social Council (ECOSOC) adopted a resolution during its substantive session to establish the United Nations Committee of Experts on Global Geospatial Information Management (UN-GGIM). This Committee of Experts is an official intergovernmental body composed of government experts, typically from national geospatial information authorities of UN Member States, as well as experts from competent international organizations participating as observers.
Some of the main objectives of the Committee of Experts include: a) discussing, enhancing, and coordinating global geospatial information management activities by engaging Member States at the highest levels as key participants; b) making joint decisions and agreeing on guidelines for the use of geospatial information within national and global policy and regulatory frameworks; c) working with governments to improve policies, institutional arrangements, and legal frameworks, addressing global challenges, and contributing to collective knowledge as a community of shared interests and concerns; and d) developing effective strategies to build geospatial capacity in developing countries. UN-GGIM is divided into five regional committees: Asia-Pacific, Americas, Europe, Arab States, and Africa.
Within the framework of the UN-GGIM Committee of Experts, High-Level Forums (UN-HLF) are held to promote comprehensive dialogue on global geospatial information management with governments, non-governmental organizations, and the private sector. These forums aim to significantly enhance and strengthen the national geospatial information management capacities of countries, particularly in developing nations.
The Government of Mexico, through the National Institute of Statistics and Geography, recognized internationally as a strong leader for its substantial technical contributions to advancing the collection, management, dissemination, and use of geospatial information, has offered to host the Seventh High-Level Forum on United Nations Global Geospatial Information Management from October 8 to 10, 2024. The Seventh High-Level Forum provides a unique opportunity for the Committee of Experts, as a connected global community, to share new ideas, strategies, approaches, technologies, and innovations to support countries, regional and global agendas, and their respective priorities, programs, and national projects.
The Seventh High-Level Forum, under the theme “Accelerating Implementation; Achieving Resilience,” will bring together high-level government participants and stakeholders to expedite the implementation of the United Nations Integrated Geospatial Information Framework to achieve the Sustainable Development Goals and address climate-related challenges.
Agenda:
- Presentation of the ITU GeoAI Challenge: goals, objectives and outlook. Andrea Manara, ITU and UN Geospatial Network
- Presentations of the problem statements, moderated by Maria Antonia Brovelli, UN GGIM Academic Network
- First challenge, Vegetation Mapping Challenge presenter(s) Elio Villaseñor and José Luis Ornelas (INEGI)
- Q&A
- Human Settlement Detection Challenge, presenter(s) Elio Villaseñor and Rodrigo Suárez (INEGI)
- Q&A
- Outlook. Maria Antonia Brovelli, Politecnico di Milano
- Q&A general
