P-SLCR: Unsupervised Point Cloud Semantic Segmentation via Prototypes Structure Learning and Consistent Reasoning

2026-03-06

AI Summary

This work addresses the challenges of unsupervised point cloud semantic segmentation, where the absence of annotations and pretraining support hinders performance. To overcome these limitations, we propose a novel prototype-based approach that leverages consistent structural learning and semantic relational consistency reasoning to effectively establish semantic associations between points and prototypes. The method enforces high-quality feature constraints to maintain consistency between crisp and ambiguous prototype banks. Key innovations include a prototype-bank-driven strategy, modeling of inter-prototype relational matrices, and an unsupervised semantic reasoning mechanism. Extensive experiments demonstrate state-of-the-art performance on S3DIS, SemanticKITTI, and ScanNet benchmarks, with a remarkable 47.1% mIoU on S3DIS Area-5—surpassing the fully supervised PointNet for the first time in this setting.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Semi-Supervised Learning
• Computer Vision: Segmentation

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Current semantic segmentation approaches for point cloud scenes heavily rely on manual labeling, while research on unsupervised semantic segmentation methods specifically for raw point clouds is still in its early stages. Unsupervised point cloud learning poses significant challenges due to the absence of annotation information and the lack of pre-training. The development of effective strategies is crucial in this context. In this paper, we propose a novel prototype library-driven unsupervised point cloud semantic segmentation strategy that utilizes Structure Learning and Consistent Reasoning (P-SLCR). First, we propose a Consistent Structure Learning to establish structural feature learning between consistent points and the library of consistent prototypes by selecting high-quality features. Second, we propose a Semantic Relation Consistent Reasoning that constructs a prototype inter-relation matrix between consistent and ambiguous prototype libraries separately. This process ensures the preservation of semantic consistency by imposing constraints on consistent and ambiguous prototype libraries through the prototype inter-relation matrix. Finally, our method was extensively evaluated on the S3DIS, SemanticKITTI, and Scannet datasets, achieving the best performance compared to unsupervised methods. Specifically, the mIoU of 47.1% is achieved for Area-5 of the S3DIS dataset, surpassing the classical fully supervised method PointNet by 2.5%.

Problem

• unsupervised semantic segmentation
• point cloud
• annotation-free learning
• prototype learning
• semantic consistency

Innovation

• Unsupervised Semantic Segmentation
• Point Cloud
• Prototype Learning
• Structure Learning
• Consistent Reasoning