发表机构
School of Automation, Beijing Institute of Technology; Zhongguancun Academy; Artificial Intelligence Academy, Xidian University; School of Information and Communication Engineering, Beijing University of Posts and Telecommunications（北京理工大学自动化学院; 中关村学院; 西安电子科技大学人工智能学院; 北京邮电大学信息与通信工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
GRC-Pose提出生成-重建对应框架，结合GeoCorr-Matcher与FGH-Solver，实现无先验6D物体姿态跟踪，在HOT3D上运动保持提升58%，并保持经典基准竞争力。
AI 中文摘要
无先验6D物体姿态跟踪旨在从单个RGB视频中恢复未见物体的轨迹，而无需特定物体的CAD模型、姿态参考图像或姿态标注。几何基础模型提供了互补的物体中心和场景中心线索，然而SAM3D CAD以任意物体局部表面参数化进行索引，而重建证据则以具有部分表面覆盖的序列特定世界坐标系表达。为利用这种互补性，我们将跟踪表述为生成-重建对应，并引入GRC-Pose，一种基于对应的框架，结合了学习对应预测与鲁棒姿态估计。具体而言，GeoCorr-Matcher为每个姿态候选估计加权物体-场景对应及每匹配的不确定性。FGH-Solver通过多个鲁棒几何估计器和序列级后验推理整合这些匹配，而后验门控记忆仅保留通过遮挡和视角变化的内点支持的观测。广泛评估表明，使用SAM3D CAD，GRC-Pose在HOT3D上实现了最先进的平均召回率和运动保持，后者较先前技术提升58%。在包括YCBInEOAT和LINEMOD的经典基准上，它仍保持高度竞争力。
英文摘要
Prior-free 6D object pose tracking seeks to recover the trajectory of an unseen object from a single RGB video without object-specific CAD models, posed reference images, or pose annotations. Geometric foundation models provide complementary object-centric and scene-centric cues, yet SAM3D CAD is indexed by an arbitrary object-local surface parameterization, whereas reconstructed evidence is expressed in a sequence-specific world frame with partial surface coverage. To exploit this complementarity, we formulate tracking as generation-reconstruction correspondence and introduce GRC-Pose, a correspondence-based framework that combines learned correspondence prediction with robust pose estimation. Concretely, GeoCorr-Matcher estimates weighted object-scene correspondences and per-match uncertainty for each pose candidate. FGH-Solver integrates these matches through multiple robust geometric estimators and sequence-level posterior inference, while a posterior-gated memory retains only inlier-supported observations through occlusion and viewpoint change. Extensive evaluation shows that with SAM3D CAD, GRC-Pose achieves state-of-the-art Average Recall and motion retention on HOT3D, improving the latter by 58% over prior art. On classical benchmarks including YCBInEOAT and LINEMOD, it remains highly competitive.
Comments39 pages
