TreeLearn: A deep learning method for segmenting individual trees from ground-based LiDAR forest point clouds

Jonathan Henrich, Jan van Delden, Dominik Seidel, Thomas Kneib, Alexander S. Ecker

Abstract: Highlights•A deep learning method for tree extraction operating directly on forest point clouds.•Training can be supervised with data generated by existing segmentation algorithms.•Fine-tuning the network with high-quality data substantially improves performance.•We provide a manually segmented forest point cloud for training and evaluation.