Contrastive Attention-Based Network for Self-Supervised Point Cloud Completion

Seema Kumari

doi:10.1109/LSP.2025.3631424

Contrastive Attention-Based Network for Self-Supervised Point Cloud Completion

2025-01-01

0

PRE

AI

DOI：10.1109/LSP.2025.3631424

Original

Original request for help

Save

Abstract

En 中文

Point cloud completion aims to reconstruct complete 3D shapes from partial observations, often requiring multiple views or complete data for training. In this paper, we propose an attention-driven, self-supervised autoencoder network that completes 3D point clouds from a single partial observation. Multi-head self-attention captures robust contextual relationships, while residual connections in the autoencoder enhance geometric feature learning. In addition to this, we incorporate a contrastive learning-based loss, which encourages the network to better distinguish structural patterns even in highly incomplete observations. Experimental results on benchmark datasets demonstrate that the proposed approach achieves state-of-the-art performance in single-view point cloud completion.

Keywords:

Auto-encoder

multi-head self-attention

point cloud completion

contrastive learning

and self-supervision

Contrastive Attention-Based Network for Self-Supervised Point Cloud Completion

Abstract

Journal

Organization

Cited Papers