Who Handles Orientation? Investigating Invariance in Feature Matching

[论文] Who Handles Orientation? Investigating Invariance in Feature Matching ## 论文概要 **研究领域**: cs.CV **作者**: David Nordström, Johan Edstedt, Fredrik Kahl, Georg Bökman **发布时间**: 2026-04-13 **arXiv**: [2604.11809](https://arxiv.org/abs/2604.11809) ## 中文摘要 图像间匹配关键点是3D计算机视觉的核心问题。现代匹配器在处理大平面旋转时存在困难。本文研究了旋转不变性应该在哪个阶段被引入，发现将旋转不变性整合到描述符中已经能获得与在匹配器中处理相似的性能，但能更快地实现旋转不变性。研究还表明，在训练规模足够大时，强制旋转不变性不会影响正常方向的性能。我们发布了两个对平面旋转具有鲁棒性的匹配器，在多模态、极端场景和卫星图像匹配上达到了最先进的性能。 ## 原文摘要 Finding matching keypoints between images is a core problem in 3D computer vision. However, modern matchers struggle with large in-plane rotations. A straightforward mitigation is to learn rotation invariance via data augmentation. However, it remains unclear at which stage rotation invariance should be incorporated. In this paper, we study this in the context of a modern sparse matching pipeline. We perform extensive experiments by training on a large collection of 3D vision datasets and evaluating on popular image matching benchmarks. Surprisingly, we find that incorporating rotation invariance already in the descriptor yields similar performance to handling it in the matcher. However, rotation invariance is achieved earlier in the matcher when it is learned in the descriptor, allowing for a faster rotation-invariant matcher. Further, we find that enforcing rotation invariance does not hurt upright performance when trained at scale. Finally, we study the emergence of rotation invariance through scale and find that increasing the training data size substantially improves generalization to rotated images. We release two matchers robust to in-plane rotations that achieve state-of-the-art performance on e.g. multi-modal (WxBS), extreme (HardMatch), and satellite image matching (SatAst). Code is available at https://github.com/davnords/loma. --- *自动采集于 2026-04-15* #论文 #arXiv #AI #小凯