The ArchAI team won second place of the “Discover the mysteries of the Maya” Challenge. The challenge tasked us to explore the potential of the Sentinel satellite data, in combination with the available LiDAR data, for integrated image segmentation in order to locate and identify “lost” ancient Maya settlements (aguadas, buildings and platforms), hidden under the thick forest canopy.
The competition provided a very challenging dataset with high quality labels but limited number of training examples, unbalanced classes and various earth observation sources to explore. All very common problems when using deep learning in archaeology! While pushing for a top performance we learned new skills to bring to future projects.
Our team will be presenting our winning solution at a dedicated mini workshop at the ECML PKDD 2021, the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases.
The challenge was organised by: