PointCSP: Cross-Sample Semantic Propagation and Stability Preservation in Self-Supervised Point Cloud Learning

2026-05-03

AI Summary

This work addresses the challenge of semantic inconsistency across scenes in existing self-supervised point cloud learning methods, which typically model samples independently. To overcome this limitation, the authors propose a novel self-supervised framework based on Cross-sample Semantic Propagation (CSP). By serializing batch samples and incorporating a state space model, the method explicitly captures dynamic inter-sample dependencies to achieve semantic alignment in the latent space. Furthermore, during fine-tuning, an Asymmetric Semantic-Preserving Distillation (SPD) mechanism is introduced to mitigate semantic drift when testing on single scenes. As the first approach to integrate cross-sample semantic propagation into self-supervised point cloud representation learning, the proposed method achieves state-of-the-art performance across multiple benchmarks, demonstrating superior results in both accuracy and semantic consistency.

Abstract

Scene-level point cloud self-supervised learning (PC-SSL) has demonstrated potential in enhancing the generalization capability of 3D vision models. Despite the advances in the field through existing methods, the sample-independent modeling paradigm still poses significant limitations in terms of maintaining consistent semantic representations across scenes. This challenge hinders the construction of a unified and transferable semantic space. To address this issue, we propose a PC-SSL framework based on cross-sample semantic propagation (CSP), in which samples within a batch are serialized into continuous input and processed by a state-space model to enable semantic state propagation. This mechanism explicitly models the dynamic dependencies across samples in the state space, allowing the network to establish cross-sample semantic consistency in the latent space and achieve global semantic alignment. Since serialization-based pretraining requires batch-level input organization, we further introduce an asymmetric semantic preservation distillation (SPD) during finetuning to achieve structural alignment of semantic transfer and eliminate inconsistencies caused by batch dependency. The proposed SPD ensures stable transfer of pretrained semantics through a heterogeneous input mechanism and a semantic feature alignment constraint. This enables the model to maintain structured semantic consistency and robustness under single-scene testing conditions. Extensive experiments on multiple benchmark datasets demonstrate that our method consistently outperforms state-of-the-art methods in both performance and semantic consistency.

Problem

• self-supervised learning
• point cloud
• semantic consistency
• cross-sample modeling
• semantic representation

Innovation

• Cross-Sample Semantic Propagation
• State-Space Model
• Semantic Consistency
• Self-Supervised Point Cloud Learning
• Semantic Preservation Distillation