Cov2Pose：利用空间协方差实现流形感知的 6-DoF 物体姿态直接估计
Cov2Pose: Leveraging Spatial Covariance for Direct Manifold-aware 6-DoF Object Pose Estimation
用协方差池化把卷积特征分布编码成 SPD 矩阵、再经 Cholesky 分解做连续姿态编码，并在黎曼几何约束下回归，可直接替换现有直接姿态回归头中的全局池化层，提升遮挡下的稳定性。
原文依据 we propose a covariance-pooled representation that encodes convolutional feature distributions as a symmetric positive definite (SPD) matrix
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文研究从单张 RGB 图像估计物体 6-DoF 姿态的问题。先预测中间 2D 关键点、再用透视 n 点（PnP）求解器的间接方法已展现出优异性能。以端到端方式回归姿态的直接方法通常计算效率更高，但精度较低。然而，直接姿态回归头依赖全局池化特征，忽略了空间二阶统计量，而后者在姿态预测中颇具信息量。此外，它们在多数情况下预测的是不连续的姿态表示，缺乏鲁棒性。为此，本文提出一种协方差池化表示，将卷积特征分布编码为对称正定（SPD）矩阵。进一步地，我们通过 SPD 矩阵的 Cholesky 分解提出一种新颖的姿态编码形式。随后，借助考虑 SPD 矩阵黎曼几何的流形感知网络头，以端到端方式回归姿态。实验与消融研究一致表明，二阶池化与连续表示对直接姿态回归具有重要价值，在部分遮挡条件下亦然。
英文原文摘要 1207 字符
In this paper, we address the problem of 6-DoF object pose estimation from a single RGB image. Indirect methods that typically predict intermediate 2D keypoints, followed by a Perspective-n-Point solver, have shown great performance. Direct approaches, which regress the pose in an end-to-end manner, are usually computationally more efficient but less accurate. However, direct pose regression heads rely on globally pooled features, ignoring spatial second-order statistics despite their informativeness in pose prediction. They also predict, in most cases, discontinuous pose representations that lack robustness. Herein, we therefore propose a covariance-pooled representation that encodes convolutional feature distributions as a symmetric positive definite (SPD) matrix. Moreover, we propose a novel pose encoding in the form of an SPD matrix via its Cholesky decomposition. Pose is then regressed in an end-to-end manner with a manifold-aware network head, taking into account the Riemannian geometry of SPD matrices. Experiments and ablations consistently demonstrate the relevance of second-order pooling and continuous representations for direct pose regression, including under partial occlusion.