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

TL;DR

This paper proposes a novel prototype library-driven unsupervised point cloud semantic segmentation strategy that utilizes Structure Learning and Consistent Reasoning (P-SLCR) to establish structural feature learning between consistent points and the library of consistent prototypes by selecting high-quality features.

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%.

Figures (7)

• Figure 1: P-SLCR architecture for structural learning and consistent reasoning through a learnable prototype-based library. The consistent points are continuously expanding and increasingly resembling the prototype.
• Figure 2: An overview of our P-SLCR with features extracted by SparseConv, pseudo-labels and predictions obtained by clustering and segmenting the header. The proposed reliability classifies the scene into consistent and ambiguous points and establishes an updatable prototype library to construct consistent structure learning and semantic relation consistent reasoning based on the prototype library.
• Figure 3: The qualitative results on S3DIS demonstrate the segmentation performance in comparison to GrowSP. Each color represents a semantic class. We highlight the differences with black circles.
• Figure 4: The qualitative results on SemanticKITTI demonstrate the segmentation performance in comparison to GrowSP. We zoomed in on local regions to show details and highlighted differences with black circles.
• Figure 5: The qualitative results on ScanNet demonstrate the segmentation performance in comparison to GrowSP. Each color represents a semantic class. We highlight the differences with black circles.