Scalable Unseen Objects 6-DoF Absolute Pose Estimation with Robotic Integration
可扩展的未知物体六自由度绝对位姿估计与机器人集成
机构 * National Engineering Research Center for Robot Visual Perception and Control Technology, School of Artificial Intelligence and Robotics, Hunan University(机器人视觉感知与控制技术国家工程研究中心,人工智能与机器人学院,湖南大学) ; School of Architecture and Art, Central South University(建筑与艺术学院,中南大学) ; School of Computing and Communications, Lancaster University(计算与通讯学院,兰卡斯特大学) ; Department of Computer Science and Software Engineering, The University of Western Australia(计算机科学与软件工程系,西澳大学) ; School of Electrical and Electronic Engineering, Nanyang Technological University(电子与电气工程学院,南洋理工大学)
AI总结 本文提出SinRef-6D方法,通过单张姿态标注的RGB-D图像实现未知物体六自由度位姿估计,利用状态空间模型解决大位姿差异和单视角信息有限的问题,实验验证其在多个基准和真实场景中的优越可扩展性。
Comments Accepted by TRO 2026, 18 pages, 9 figures
Journal ref IEEE Transactions on Robotics, 2026