Benchmarking Deep Learning Models for Aerial LiDAR Point Cloud Semantic Segmentation under Real Acquisition Conditions: A Case Study in Navarre

2026-03-23

AI Summary

This study addresses the lack of systematic evaluation of deep learning models for semantic segmentation of airborne LiDAR point clouds under real-world aerial survey conditions, particularly in complex urban, rural, and industrial scenes where class imbalance and geometric diversity pose significant challenges. Leveraging a large-scale, real-flight LiDAR dataset collected over Navarre, Spain, we present the first end-to-end benchmark of four state-of-the-art 3D segmentation models—KPConv, RandLA-Net, Superpoint Transformer, and Point Transformer V3—under unified flight parameters and diverse land cover types. All models achieve overall accuracy above 93%, with KPConv attaining the highest mean IoU of 78.51%. Point Transformer V3 excels on rare classes such as vehicles, reaching 75.11% IoU, while Superpoint Transformer and RandLA-Net demonstrate superior computational efficiency, highlighting a clear trade-off between accuracy and inference speed.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Recent advances in deep learning have significantly improved 3D semantic segmentation, but most models focus on indoor or terrestrial datasets. Their behavior under real aerial acquisition conditions remains insufficiently explored, and although a few studies have addressed similar scenarios, they differ in dataset design, acquisition conditions, and model selection. To address this gap, we conduct an experimental benchmark evaluating several state-of-the-art architectures on a large-scale aerial LiDAR dataset acquired under operational flight conditions in Navarre, Spain, covering heterogeneous urban, rural, and industrial landscapes. This study compares four representative deep learning models, including KPConv, RandLA-Net, Superpoint Transformer, and Point Transformer V3, across five semantic classes commonly found in airborne surveys, such as ground, vegetation, buildings, and vehicles, highlighting the inherent challenges of class imbalance and geometric variability in aerial data. Results show that all tested models achieve high overall accuracy exceeding 93%, with KPConv attaining the highest mean IoU (78.51%) through consistent performance across classes, particularly on challenging and underrepresented categories. Point Transformer V3 demonstrates superior performance on the underrepresented vehicle class (75.11% IoU), while Superpoint Transformer and RandLA-Net trade off segmentation robustness for computational efficiency.

Problem

• aerial LiDAR
• semantic segmentation
• deep learning
• benchmarking
• real acquisition conditions

Innovation

• aerial LiDAR
• semantic segmentation
• deep learning benchmark
• real-world acquisition
• point cloud