The New Zealand Point Cloud Classification Deep Learning Package is designed to classify point clouds into two categories: trees and background. Optimized specifically for New Zealand aerial LiDAR data, this model delivers high accuracy in identifying tree structures across both urban and rural environments.

Classifying trees from point cloud data is valuable for a wide range of applications, including 3D basemap generation, urban planning, forestry management, and climate change response initiatives. Due to the complex and irregular geometry of trees, traditional methods often fall short. Deep learning models, however, excel at capturing these intricate patterns, resulting in more accurate classifications.

This model has been trained, tested, and validated using datasets collected across New Zealand, ensuring strong performance in recognizing vegetation typical to the region.