#Machine-Learning
TreeGPT: Generative Pre-trained Transformer for forestry applications with 3D point clouds
For my master's thesis, I tackled a challenge that has captured my interest for years: leveraging LiDAR point clouds to generate automated forest inventories. As of today, we still rely on largely incomplete manual forest inventories to manage the wooden third of Switzerland‘s land area. Advancements in 3D computer vision applied to LiDAR point clouds can improve that. TreeGPT leverages self-supervised training on unlabeled and synthetic point clouds of single trees and contributes to the creation of a foundation model for forestry vision tasks.
TreeGPT: Generative Pre-trained Transformer for forestry applications with 3D point cloudsPre-print: Automatic inventory of retaining walls from aerial lidar data using 3D deep learning
Infrastructure management along highways and railways requires inventories of critical structures like retaining walls, which traditionally relies on manual inspection and documentation. Unfortunately, data in infrastructure databases is often incomplete. In this study, we investigated the feasibility of automating retaining wall inventories using public aerial lidar data from the Swiss Federal Office of Topography (Swisstopo).
Pre-print: Automatic inventory of retaining walls from aerial lidar data using 3D deep learning