论文标题
LM-RELOC：基于Levenberg-Marquardt直接视觉重新定位
LM-Reloc: Levenberg-Marquardt Based Direct Visual Relocalization
论文作者
论文摘要
我们提出LM-RELOC - 一种基于直接图像比对的视觉重新定位的新方法。与以基于功能的配方解决问题的先前作品相反，该方法不依赖于功能匹配和RANSAC。因此，该方法不仅可以使用梯度，而且可以利用图像的任何区域。特别是，我们提出了一种受莱文伯格·马尔夸特（Levenberg-Marquardt）算法训练LM-NET的启发的损失配方。学到的特征显着提高了直接图像对齐的鲁棒性，尤其是在不同条件下重新定位的。为了进一步提高LM-NET对大图像基线的鲁棒性，我们提出了一个姿势估计网络Corrposenet，该网络会回归相对姿势以引导直接图像比对。对Carla和牛津机器人重新定位跟踪基准的评估表明，我们的方法比以前的最先进方法提供了更准确的结果，同时在鲁棒性方面具有可比性。
We present LM-Reloc -- a novel approach for visual relocalization based on direct image alignment. In contrast to prior works that tackle the problem with a feature-based formulation, the proposed method does not rely on feature matching and RANSAC. Hence, the method can utilize not only corners but any region of the image with gradients. In particular, we propose a loss formulation inspired by the classical Levenberg-Marquardt algorithm to train LM-Net. The learned features significantly improve the robustness of direct image alignment, especially for relocalization across different conditions. To further improve the robustness of LM-Net against large image baselines, we propose a pose estimation network, CorrPoseNet, which regresses the relative pose to bootstrap the direct image alignment. Evaluations on the CARLA and Oxford RobotCar relocalization tracking benchmark show that our approach delivers more accurate results than previous state-of-the-art methods while being comparable in terms of robustness.
