CAS-NET: Cascade attention-based sampling neural network for point cloud simplification

Date

2023-07

Advisors

Journal Title

Journal ISSN

ISSN

Volume Title

Publisher

IEEE

Type

Conference

Peer reviewed

Yes

Abstract

Point cloud sampling can reduce storage requirements and computation costs for various vision tasks. Traditional sampling methods, such as farthest point sampling, are not geared towards downstream tasks and may fail on such tasks. In this paper, we propose a cascade attention-based sampling network (CAS-Net), which is end-to-end trainable. Specifically, we propose an attention-based sampling module (ASM) to capture the semantic features and preserve the geometry of the original point cloud. Experimental results on the ModelNet40 dataset show that CAS-Net outperforms state-of-the-art methods in a sampling-based point cloud classification task, while preserving the geometric structure of the sampled point cloud.

Description

Keywords

Point clouds, Attention-based sampling

Citation

Chen, C., Yuan, H.. Liu, H., Hou, J. and Hamzaoui, R. (2023) CAS-NET: Cascade attention-based sampling neural network for point cloud simplification, Proc. IEEE ICME 2023, Brisbane, July 2023.

Rights

Research Institute

Institute of Engineering Sciences (IES)