西华大学学报（自然科学版）2025，Vol.44Issue(6)：82-90,9.DOI:10.12198/j.issn.1673-159X.5530
基于RGB-D数据的改进PVN3D的6D位姿估计算法
6D Pose Estimation Algorithm Based on RGB-D Data and Improved PVN3D
摘要
Abstract
In the fields of computer vision and robotics,6D pose estimation is an important task.Giv-en that existing methods for 6D pose estimation based on RGB-D images struggle to fully utilize feature in-formation.This paper proposes an improved 6D pose estimation algorithm.The algorithm leverages the ad-vantages of the YOLOv8n-seg and ResNet-UNet frameworks to effectively extract and utilize multimodal information from both RGB images and point cloud data.Semantic segmentation of RGB images is achieved using the YOLOv8n-seg module based on the PVN3D network,this method captures more de-tailed scene features.Additionally,the introduction of ResNet-UNet enhances detection accuracy through feature cascading and multiscale information fusion.Customized optimization of the loss function further improves overall performance.Experimental results show that on the LineMOD dataset,the mean precision is improved by 2%on 13 different object types,which validated the effectiveness of the proposed im-proved algorithm.关键词
6D位姿估计/RGB-D/图像处理/YOLOv8Key words
6D pose estimation/RGB-D/image processing/YOLOv8分类
信息技术与安全科学引用本文复制引用
张筱晨,刘建新,黄天才,陈博..基于RGB-D数据的改进PVN3D的6D位姿估计算法[J].西华大学学报（自然科学版）,2025,44(6):82-90,9.基金项目
教育部春晖计划项目(Z2017084). （Z2017084）
