Stefano Puliti

The paper introduces MillionTrees, a global benchmark dataset for airborne tree detection, showing that current models struggle to generalize across diverse data and that training on MillionTrees improves performance.

This paper develops a deep learning method to reconstruct coniferous tree crown shapes from incomplete point clouds, improving accuracy in structural metrics.

The paper introduces deadtrees.earth, an open-access database of high-resolution aerial imagery to map global tree mortality using community-sourced data and machine learning.