HOLO：面向SD地图细粒度视觉定位的单应性引导位姿估计网络
HOLO: Homography-Guided Pose Estimator Network for Fine-Grained Visual Localization on SD Maps
将单应性约束引入多视图图像与SD地图的跨模态定位，通过BEV投影和语义对齐构建满足单应性的输入对，并利用单应关系引导特征融合与位姿回归，为自动驾驶低成本定位提供了可复用的几何先验注入方式。
原文依据 We construct input pairs that satisfy a homography constraint by projecting ground-view features into the BEV domain and enforcing semantic alignment with map features.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于标准清晰度（SD）地图的视觉定位已成为自动驾驶中一种有前景的低成本且可扩展的解决方案。然而，现有的基于回归的方法往往忽略了固有的几何先验，导致训练效率次优且定位精度有限。本文提出了一种新颖的单应性引导位姿估计网络，用于多视图图像与标准清晰度（SD）地图之间的细粒度视觉定位。我们通过将地面视图特征投影到鸟瞰图（BEV）域并强制与地图特征进行语义对齐，构建满足单应性约束的输入对。然后，我们利用单应关系来引导特征融合，并将位姿输出限制在有效的可行区域内，与依赖基于注意力的融合和直接3自由度位姿回归的先前方法相比，这显著提高了训练效率和定位精度。据我们所知，这是首个统一BEV语义推理与单应性学习用于图像到地图定位的工作。此外，通过显式建模单应变换，所提出的框架自然支持跨分辨率输入，增强了模型灵活性。在nuScenes数据集上的大量实验表明，我们的方法显著优于现有的最优视觉定位方法。代码和预训练模型将在https://github.com/Quartararo0714/HOLO公开。
英文原文摘要 1491 字符
Visual localization on standard-definition (SD) maps has emerged as a promising low-cost and scalable solution for autonomous driving. However, existing regression-based approaches often overlook inherent geometric priors, resulting in suboptimal training efficiency and limited localization accuracy. In this paper, we propose a novel homography-guided pose estimator network for fine-grained visual localization between multi-view images and standard-definition (SD) maps. We construct input pairs that satisfy a homography constraint by projecting ground-view features into the BEV domain and enforcing semantic alignment with map features. Then we leverage homography relationships to guide feature fusion and restrict the pose outputs to a valid feasible region, which significantly improves training efficiency and localization accuracy compared to prior methods relying on attention-based fusion and direct 3-DoF pose regression. To the best of our knowledge, this is the first work to unify BEV semantic reasoning with homography learning for image-to-map localization. Furthermore, by explicitly modeling homography transformations, the proposed framework naturally supports cross-resolution inputs, enhancing model flexibility. Extensive experiments on the nuScenes dataset demonstrate that our approach significantly outperforms existing state-of-the-art visual localization methods. Code and pretrained models will be publicly available at https://github.com/Quartararo0714/HOLO.