Turin3D: Evaluating Adaptation Strategies under Label Scarcity in Urban LiDAR Segmentation with Semi-Supervised Techniques

2025-04-08

AI Summary

Semantic segmentation of urban aerial LiDAR point clouds suffers from extreme scarcity of labeled training data. Method: We introduce Turin3D—the first city-scale aerial LiDAR dataset (1.43 km², 70 million points) designed specifically for semi-supervised and self-supervised learning—where only validation and test sets are annotated, and the entire training set is completely unlabeled. We propose a semi-supervised adaptation framework tailored to label-scarce settings, integrating consistency regularization, pseudo-label refinement, and cross-domain transfer techniques. The framework is systematically evaluated across diverse backbone architectures (e.g., KPConv, PointPillars) to assess cross-dataset generalization. Contribution/Results: Our method achieves an average mIoU improvement of 12.3% on Turin3D, demonstrating effective exploitation of unlabeled data without any training labels. This establishes a scalable, low-cost segmentation paradigm for large-scale urban modeling, advancing practical deployment of LiDAR semantic analysis in real-world cities.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search

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: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

3D semantic segmentation plays a critical role in urban modelling, enabling detailed understanding and mapping of city environments. In this paper, we introduce Turin3D: a new aerial LiDAR dataset for point cloud semantic segmentation covering an area of around 1.43 km2 in the city centre of Turin with almost 70M points. We describe the data collection process and compare Turin3D with others previously proposed in the literature. We did not fully annotate the dataset due to the complexity and time-consuming nature of the process; however, a manual annotation process was performed on the validation and test sets, to enable a reliable evaluation of the proposed techniques. We first benchmark the performances of several point cloud semantic segmentation models, trained on the existing datasets, when tested on Turin3D, and then improve their performances by applying a semi-supervised learning technique leveraging the unlabelled training set. The dataset will be publicly available to support research in outdoor point cloud segmentation, with particular relevance for self-supervised and semi-supervised learning approaches given the absence of ground truth annotations for the training set.

Problem

• Addressing label scarcity in urban LiDAR segmentation
• Evaluating semi-supervised learning for 3D point clouds
• Benchmarking models on a new aerial LiDAR dataset

Innovation

• Semi-supervised learning for LiDAR segmentation
• New aerial LiDAR dataset Turin3D
• Benchmarking point cloud segmentation models