arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.02008cs.CV
收藏

GlueStick:通过将点和线粘合在一起实现鲁棒图像匹配

GlueStick: Robust Image Matching by Sticking Points and Lines Together

ETH Zurich(苏黎世联邦理工学院);Qualcomm XR Labs Europe(高通欧洲XR实验室);Microsoft Mixed Reality and AI Zurich lab(微软苏黎世混合现实与人工智能实验室);Lund University(隆德大学)

Rémi Pautrat, Iago Suárez, Yifan Yu, Marc Pollefeys, Viktor Larsson

AI总结:

本文提出了一种名为 GlueStick 的深度匹配图神经网络,通过将点、线及其描述子统一到单一的线框结构中,利用节点间的连接信息实现联合匹配,显著提升了图像匹配的鲁棒性和性能。

摘要:

线段是与点互补的强大特征。它们提供结构线索,对剧烈的视角和光照变化具有鲁棒性,并且即使在无纹理区域中也能存在。然而,由于部分遮挡、缺乏纹理或重复性,与点相比,描述和匹配线段更具挑战性。本文引入了一种新的匹配范式,将点、线及其描述子统一到单一的线框结构中。我们提出了 GlueStick,一种深度匹配图神经网络(GNN),它接收来自不同图像的两个线框,并利用节点之间的连接信息将它们更好地粘合在一起。除了联合匹配带来的效率提升外,我们还展示了在单一架构中利用这两种特征互补性所带来的巨大性能提升。我们表明,对于广泛的数据集和任务,我们的匹配策略优于独立匹配线段和点的最先进方法。代码可在 https://github.com/cvg/GlueStick 获取。

英文摘要:

Line segments are powerful features complementary to points. They offer structural cues, robust to drastic viewpoint and illumination changes, and can be present even in texture-less areas. However, describing and matching them is more challenging compared to points due to partial occlusions, lack of texture, or repetitiveness. This paper introduces a new matching paradigm, where points, lines, and their descriptors are unified into a single wireframe structure. We propose GlueStick, a deep matching Graph Neural Network (GNN) that takes two wireframes from different images and leverages the connectivity information between nodes to better glue them together. In addition to the increased efficiency brought by the joint matching, we also demonstrate a large boost of performance when leveraging the complementary nature of these two features in a single architecture. We show that our matching strategy outperforms the state-of-the-art approaches independently matching line segments and points for a wide variety of datasets and tasks. The code is available at https://github.com/cvg/GlueStick.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑