Sparse Anchors for Visual Multi-Session Registration in Changing Environments
作者: Yunus Talha Erzurumlu, Niklas Trekel, Saurabh Gupta, Nico Koltermann, Olga Vysotska, Jens Behley, Louis Wiesmann, Cyrill Stachniss
分类: cs.CV, cs.RO
发布日期: 2026-11-03
备注: 8 pages, 4 figures, 2 tables
💡 一句话要点
提出稀疏锚点以解决变化环境中的视觉多会话注册问题
🎯 匹配领域: 支柱六:视频提取与匹配 (Video Extraction)
关键词: 视觉多会话注册 稀疏锚点 机器人监测 环境变化 姿态优化
📋 核心要点
- 现有方法在环境变化时难以保持准确的跨会话定位,导致视觉内容减少和错误对应。
- 本研究通过稀疏绝对姿态约束与相对姿态信息结合,提出了一种新的视觉多会话注册方法。
- 在HILTI26数据集上进行的实验显示,该方法在1米和5米内的回调率分别达到94.4%和100.0%,显著优于现有方法。
📝 摘要(中文)
重复的机器人监测需要在环境变化的情况下进行定位。环境的变化可能显著减少持久的视觉内容,而重复结构可能导致错误的跨会话对应关系。本文提出了一种视觉多会话注册方法,将新的数据记录(查询会话)与先前重建的参考地图进行对齐。在显著的视觉和结构变化下,跨会话定位可能仅对部分轨迹有效。我们依赖一组稀疏且可靠的绝对姿态约束,并结合会话内的相对姿态信息来完成完整轨迹的注册。通过视觉位置识别、局部特征匹配和几何验证,我们首先估计绝对对齐假设,然后使用序列共识来拒绝不一致的假设,保留剩余的姿态作为绝对锚点。经过高斯特征场的特征度量对齐后,这些锚点被优化,并与会话内的相对姿态约束结合。我们的实验表明,该方法在HILTI26建筑工地数据集上表现优异,回调率达到94.4%。
🔬 方法详解
问题定义:本文旨在解决在环境变化情况下的视觉多会话注册问题。现有方法在处理视觉内容减少和重复结构时,容易导致错误的跨会话对应关系。
核心思路:我们提出通过稀疏且可靠的绝对姿态约束,结合会话内的相对姿态信息,来实现完整轨迹的注册。这种设计旨在提高在变化环境中的定位准确性。
技术框架:整体方法包括几个主要模块:首先通过视觉位置识别和局部特征匹配估计绝对对齐假设;然后使用几何验证和序列共识来筛选假设,保留有效的绝对锚点;最后通过特征度量对齐和姿态图优化来完成最终的轨迹注册。
关键创新:本研究的创新点在于引入稀疏绝对锚点的概念,能够在仅有部分可靠对应关系的情况下,仍然实现准确的轨迹注册。这与传统方法依赖于全局一致性假设的方式有本质区别。
关键设计:在方法中,采用了高斯特征场进行特征度量对齐,并通过优化算法结合会话内的相对姿态约束,确保了注册过程的鲁棒性和准确性。
🖼️ 关键图片
📊 实验亮点
实验结果表明,所提方法在HILTI26数据集上实现了94.4%的回调率(1米内)和100.0%(5米内),显著优于独立图像定位的65.3%和70.7%。这种提升展示了稀疏绝对锚点在复杂环境下的有效性。
🎯 应用场景
该研究的潜在应用领域包括机器人监测、自动驾驶、增强现实等场景,尤其是在动态和变化环境中。通过提高跨会话定位的准确性,能够显著提升机器人在复杂环境中的自主导航和任务执行能力,具有重要的实际价值和未来影响。
📄 摘要(原文)
Repeated robotic monitoring requires localization across visits to the same environment, even if the environment changes. Such changes can substantially reduce persistent visual content, and repetitive structures can lead to incorrect cross-session correspondences. In this paper, we address visual multi-session registration of a new data recording (the query session) to a previously reconstructed reference map. Under substantial visual and structural changes, cross-session localization may be possible only for parts of the trajectory. In our approach, we therefore rely on a sparse set of reliable absolute pose constraints and combine them with within-session relative-pose information to register the complete trajectory to the reference map. We first estimate absolute alignment hypotheses through visual place recognition, local feature matching, and geometric verification. We then use sequence consensus to reject hypotheses that are inconsistent with a common query-to-reference alignment and retain the remaining poses as absolute anchors. These anchors are refined through featuremetric alignment against a Gaussian feature field and combined with within-session relative-pose constraints in a robust pose-graph optimization. We evaluate our approach on 15 cross-session experiments from the challenging HILTI26 construction-site dataset, covering substantial scene changes and difficult-to-match environments. Our method achieves 94.4% recall within 1m and 100.0% within 5m, outperforming independent image localization, which peaks at 65.3% and 70.7%, respectively, as well as direct registration of independently reconstructed Structure-from-Motion geometry. Our results show that sparse, reliable absolute anchors can provide accurate and complete trajectory registration even when reliable cross-session correspondences are available only for part of the query session.