A Channel-splited Based Dual-branch Block for 3D Point Cloud Processing

Zhong Geng-jun; Li Dong

doi:10.12052/gdutxb.220145

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

A Channel-splited Based Dual-branch Block for 3D Point Cloud Processing

更新时间：2026-05-27

- A Channel-splited Based Dual-branch Block for 3D Point Cloud Processing
- Vol. 40, Issue 4, Pages: 18-23(2023)
- 作者机构：
  
  广东工业大学自动化学院,广东,广州,510006
- 作者简介：
- 基金信息：
- DOI：10.12052/gdutxb.220145
  CLC： TP391.4;TP181
- Online First：25 July 2023，
  
  Published：2023
- 稿件说明：
移动端阅览
钟耿君, 李东. 基于通道分离机制的双分支点云处理网络[J]. 广东工业大学学报, 2023, 40(4): 18-23.

Zhong Geng-jun, Li Dong. A Channel-splited Based Dual-branch Block for 3D Point Cloud Processing[J]. 2023, 40(4): 18-23.
钟耿君, 李东. 基于通道分离机制的双分支点云处理网络[J]. 广东工业大学学报, 2023, 40(4): 18-23. DOI： 10.12052/gdutxb.220145.

Zhong Geng-jun, Li Dong. A Channel-splited Based Dual-branch Block for 3D Point Cloud Processing[J]. 2023, 40(4): 18-23. DOI： 10.12052/gdutxb.220145.

摘要

本文在PointMLP方法的基础上，提出了基于通道分离机制的双分支网络模块(Channel-splited Based Dual-branch Block，CDBlock) 。CDBlock将输入特征在通道维度上切分成两组特征，并将它们输入到双分支网络模块中的不同网络分支上。具体地，双分支网络模块包括轻量网络分支和深层网络分支。轻量网络分支由残差多层感知机 (Multi-layer Perceptron

MLP) 结构组成，负责提取浅层特征信息。深层网络分支由瓶颈网络结构组成，负责提取深层语义信息。CDBlock的引入提升了网络的对点云数据的特征提取能力和学习能力，有效地提高了模型的鲁棒性。本文方法在点云分类数据集ScanObjectNN上进行了验证，总体精度和类别平均精度分别达到了86.2%和84.97%，优于PointMLP。此外，本文方法在点云分割数据集ShapeNetPart上也取得了具有竞争力的结果。相比于PointMLP，本文方法在使用更少的参数量和计算量情况下取得了更优异的结果。

Abstract

In this paper

we propose a channel-split-based dual-branch block (CDBlock) based on the PointMLP method

which splits input features into two groups along the channel dimension and feeds them into different branches of the dual-branch network module. Specifically

our proposed CDBlock consists of a lightweight branch and a heavyweight branch. The lightweight branch

which uses the structure of the residual MLP

is designed to extract coarse semantic features. The heavyweight branch

which uses the bottleneck network as backbone

is responsible for extracting more deeper and distinguishable information. By doing this

our proposed CDBlock effectively improves the network's feature extraction and learning capabilities. Experimental results show that our proposed method outperforms the existing PointMLP by achieving an overall accuracy of 86.2% and a class average accuracy of 84.97% on ScanObjectNN dataset. In particular

our approach achieves better results with using fewer parameters and computational cost than the PointMLP. Additionally

our approach also achieves encouraging performance on the ShapeNetPart dataset.

关键词

Keywords

references

QI C R, SU H, MO K, et al . Pointnet: deep learning on point sets for 3D classification and segmentation[C ] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI: IEEE, 2017: 652-660.

QI C R, YI L, SU H, et al . Pointnet++: deep hierarchical feature learning on point sets in a metric space [J ] . Advances in Neural Information Processing Systems, 2017, 30: 5099-5108.

THOMAS H, QI C R, DESCHAUD J E, et al . Kpconv: flexible and deformable convolution for point clouds[C ] //Proceedings of the IEEE/CVF International Conference on Computer Vision. Long Beach: IEEE, 2019: 6411-6420.

WU W, QI Z, FUXIN L. Pointconv: deep convolutional networks on 3D point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seoul: IEEE, 2019: 9621-9630.

XU M, DING R, ZHAO H, et al . Paconv: position adaptive convolution with dynamic kernel assembling on point clouds[C ] //Proceedings of the IEEE/ CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2021: 3173-3182.

RAN H, LIU J, WANG C. Surface representation for point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 18942-18952.

LIU Z, HU H, CAO Y, et al . A closer look at local aggregation operators in point cloud analysis[C ] // European Conference on Computer Vision. Glasgow: ECCV, 2020: 326-342.

DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16×16 words: transformers for image recognition at scale[C]//International Conference on Learning Representations. Cambridge: ICLR, 2021: 1-6.

ZHAO H, JIANG L, JIA J, et al . Point transformer[C ] //Proceedings of the IEEE/CVF International Conference on Computer Vision. Montreal: IEEE, 2021: 16259-16268.

GUO M H, CAI J X, LIU Z N, et al . Pct: point cloud transformer [J ] . Computational Visual Media, 2021, 7(2): 187-199.

XU MA, CAN Q, HAOXUAN Y, et al. Rethinking network design and local geometry in point cloud: a simple residual MLP framework[C]//International Conference on Learning Representations. Kigali: ICLR, 2022: 1-9.

ZHANG H, WU C, ZHANG Z, et al . Resnest: split-attention networks[C ] //Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 2736-2746.

SZEGEDY C, LIU W, JIA Y, et al . Going deeper with convolutions[C ] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston: IEEE, 2015: 1-9.

HE K, ZHANG X, REN S, et al . Deep residual learning for imag e recognition[C ] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.

UY M A, PHAM Q H, HUA B S, et al . Revisiting point cloud classification: a new benchmark dataset and classification model on real-world data[C ] //Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 1588-1597.

YI L, KIM V G, CEYLAN D, et al . A scalable active framework for region annotation in 3D shape collections [J ] . ACM Transactions on Graphics(ToG), 2016, 35(6): 1.

LI Y, BU R, SUN M, et al . Pointcnn: convolution on x -transformed points [J ] . Advances in Neural Information Processing Systems, 2018, 31: 1-11.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Residual Neural Network with Voting for 3D Object Detection in Point Clouds

Dynamic Adaptive Enhancement for Power Scene Images DA-IOD Algorithm for Quantitative Target Detection

Deep Subdomain Adaptation Method Based on Domain Invariant Discriminant Feature Learning

A Hotspot Detector Based on Active Learning and Visual State Space Models

Research and Implementation of a Pulsed Laser Ranging Method Based on Digital Mixing

Related Author

Yang Ji-sheng

Zhang Yun

Xu Fangyuan

Chen Jianhua

Yuan Haoliang

Chen Hailin

Hoon Heo

Feng Yongjin

Related Institution

Samsung Electro-mechanics Co., Ltd., Suwon City

School of Computer Science and Technology, Guangdong University of Technology

School of Integrated Circuits, Guangdong University of Technology

Faculty of Applied Sciences, Macao Polytechnic University, Macao

School of Computer Science of Technology, Guangdong University of Technology

AI问答

⁰