TreeON: Reconstructing 3D Tree Point Clouds from Orthophotos and Heightmaps
2026-03-11 • Graphics
Graphics
AI summaryⓘ
The authors introduce TreeON, a new AI method that creates detailed 3D models of trees using just a single aerial photo and a height map. Their approach cleverly teaches the AI using synthetic tree data and special losses based on shadows and shapes, without needing extra information like tree species or ground scans. Tests show their method works better than others and can be used to make realistic tree models for 3D maps. They also share their code and data for others to use.
3D point cloudorthophotoDigital Surface Model (DSM)neural networkssynthetic datasetgeometric supervisionshadow losssilhouette lossprocedural modelingdigital 3D maps
Authors
Angeliki Grammatikaki, Johannes Eschner, Pedro Hermosilla, Oscar Argudo, Manuela Waldner
Abstract
We present TreeON, a novel neural-based framework for reconstructing detailed 3D tree point clouds from sparse top-down geodata, using only a single orthophoto and its corresponding Digital Surface Model (DSM). Our method introduces a new training supervision strategy that combines both geometric supervision and differentiable shadow and silhouette losses to learn point cloud representations of trees without requiring species labels, procedural rules, terrestrial reconstruction data, or ground laser scans. To address the lack of ground truth data, we generate a synthetic dataset of point clouds from procedurally modeled trees and train our network on it. Quantitative and qualitative experiments demonstrate better reconstruction quality and coverage compared to existing methods, as well as strong generalization to real-world data, producing visually appealing and structurally plausible tree point cloud representations suitable for integration into interactive digital 3D maps. The codebase, synthetic dataset, and pretrained model are publicly available at https://angelikigram.github.io/treeON/.