Mexico’s 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 the UN Geospatial Network have launched two GeoAI challenges: the “Vegetation Mapping Challenge” and the “Human Settlement Detection Challenge.” These challenges seek innovative solutions that leverage artificial intelligence (AI) and machine learning to address real and urgent problems in geospatial information management. The winners of these challenges will be awarded at the Seventh High-level Forum on United Nations Global Geospatial Information Management (UN-GGIM), which will take place in Mexico City from October 8 to 10, 2024.
Human Settlement Detection Challenge
The “Human Settlement Detection Challenge” focuses on developing a binary classification model that can efficiently detect human settlements in satellite imagery. With the rapid growth of human settlements, accurate monitoring systems are crucial for managing this growth sustainably. The challenge is to build a model that can predict the presence of human settlements in the provided test data. Participants must submit their final model along with its predictions to determine the winner.
Vegetation Mapping Challenge
The “Vegetation Mapping Challenge” aims to develop machine learning models to improve training data for vegetation mapping. Given that vegetation is essential for biodiversity, carbon capture, soil formation, and water cycle regulation, this challenge seeks proposals that identify outliers in the data or suggest more plausible labels to improve model accuracy. Participants are expected to submit a final model with their predictions on the provided test data, including identifying outlier observations and, if possible, suggesting a label.
Both challenges present a unique opportunity for students, professionals, and experts from mapping agencies worldwide to showcase their skills and contribute to advancing the Sustainable Development Goals (SDGs).
Participation and Technical Support
ITU will provide participants with a state-of-the-art, free-of-charge computing platform, including free GPUs and CPUs, a hosted Jupyter Notebook server, a Python kernel, and pre-installed machine learning packages such as PyTorch and TensorFlow.
(Text courtesy: AI for Good)
