AI for Trees Final Presentations: Automated monitoring of tree growth on satellite and drone imagery

Thanks to drone data from the regreening initiatives of our partner ‪@Justdiggit‬ and open source satellite data, the AI for Trees engineers trained machine learning models to monitor growth of trees in landscape restoration projects in Africa.

What will you see?
1. Melanie Arp from Data Enrichment team presenting how the team tried a combination of approaches including manual and supervised labeling, existing datasets and data augmentation to annotate the smallest trees possible
2. Michele Pozzi from Drone Data Modeling team explaining how they predicted trees with RGB + height data from drone footage using DeepForest, U-net and the legacy model
3. Luis Blanche + John Lister from Satellite Data Modeling team showing how to transfer and scale the modeling work from drone imagery to satellite imagery and do tree predictions on satellite data.

This Challenge was organized in collaboration with Justdiggit, VU Amsterdam and our labeling partner cfru.it.
More info: https://www.fruitpunch.ai/challenges/...