首页|Reports from Beijing Key Laboratory Highlight Recent Research in Robotic Systems (A lightweight color and geometry feature extraction and fusion module for end- to-end 6D pose estimation)
Reports from Beijing Key Laboratory Highlight Recent Research in Robotic Systems (A lightweight color and geometry feature extraction and fusion module for end- to-end 6D pose estimation)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on ro botic systems. According to news reporting originating from Beijing, People's Re public of China, by NewsRx correspondents, research stated, "Although advancemen ts in red-green-blue-depth (RGB-D)-based six degree-of-freedom (6D) pose estimat ion methods, severe occlusion remains challenging." Financial supporters for this research include National Natural Science Foundati on of China. The news editors obtained a quote from the research from Beijing Key Laboratory: "Addressing this issue, we propose a novel feature fusion module that can effic iently leverage the color and geometry information in RGB-D images. Unlike prior fusion methods, our method employs a two-stage fusion process. Initially, we ex tract color features from RGB images and integrate them into a point cloud. Subs equently, an anisotropic separable set abstraction network-like network is utili zed to process the fused point cloud, extracting both local and global features, which are then combined to generate the final fusion features. Furthermore, we introduce a lightweight color feature extraction network to reduce model complex ity. Extensive experiments conducted on the LineMOD, Occlusion LineMOD, and YCB- Video datasets conclusively demonstrate that our method significantly enhances p rediction accuracy, reduces training time, and exhibits robustness to occlusion. "
Beijing Key LaboratoryBeijingPeople' s Republic of ChinaAsiaRobotic SystemsRobotics
NETL
NSTL
