.png){kind=icon}
返回
HECPG: Hyperbolic Embedding and Confident Patch-Guided Network for Point Cloud Matching
.png){kind=icon}
PRE
AI
DOI：10.1109/TGRS.2024.3370591.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
As a fundamental problem in photogrammetry and remote sensing, terrestrial laser scanner point cloud matching aims to seek a correspondence set that can match two partially overlapping point clouds. However, existing methods suffer from poor performance when dealing with low overlapping scenarios, and they also lack the ability to effectively utilize cross-space information. In this article, we propose a novel hyperbolic embedding and confident patch-guided (HECPG) network for point cloud matching. Our method leverages hyperbolic information to enhance feature representations and suppresses the effects of nonoverlapping regions through confidence guidance. Specifically, we first introduce hyperbolic attention, which effectively incorporates hyperbolic information into point cloud matching by leveraging the inherent hierarchical structure of point clouds. In addition, we propose a confident patch search (CPS) module that assigns a confidence score to each patch point. This helps to suppress outlier correspondences that may arise in nonoverlapping regions. Once we have obtained a set of high-precision point correspondences, we use the RANSAC algorithm to estimate the alignment transform for rigid point cloud registration. Extensive experiments on indoor, outdoor, and synthetic benchmarks demonstrate the superior performance of our HECPG.
Keyword:
Point cloud compression
Feature extraction
Transformers
Task analysis
Three-dimensional displays
Mathematical models
Benchmark testing
3-D vision
attention mechanism
hyperbolic space
point cloud matching
point cloud registration
期刊
IF:
8.6
论文数:
2.1W
被引数:
10.7W
机构
引用论文
分享收藏Types of Parent Verbal Responsiveness That Predict Language in Young Children With Autism Spectrum Disorder预测自闭症谱系障碍幼儿语言的父母言语反应类型
OAAI
OAAI分享收藏GLoCNet: Robust Feature Matching With GlobalLocal Consistency Network for Remote Sensing Image RegistrationGLoCNet: 用于遥感图像配准的具有全局一致性网络的鲁棒特征匹配
PREAI
分享收藏Subversion of Autophagy in Adherent Invasive Escherichia coli-Infected Neutrophils Induces Inflammation and Cell Death.png)
PLoS ONE
OAAI
PLoS ONEIF0
OAAI分享收藏分享收藏
查看更多内容