发现现代图像检索无需盲目重排序，用匹配验证可信度更可靠
To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition
- 用匹配结果中的内点数判断是否需要重排序
- 在饱和数据集上重排序反而降低性能
- 适合追求高鲁棒性的定位系统开发者
视觉位置识别（VPR）是计算机视觉中的关键任务，传统方法通过图像匹配对检索结果进行重排序以提升性能。然而，近年来VPR方法的显著进步使得重排序的必要性受到挑战。本文表明，现代检索系统在当前饱和的数据集上，重排序常导致性能下降。我们提出将图像匹配作为验证步骤，利用内点数量来可靠预测重排序是否有益。研究结果改变了检索流程的设计范式，为构建更鲁棒、自适应的VPR系统提供了新思路。代码已开源：https://github.com/FarInHeight/To-Match-or-Not-to-Match。
原文摘要 · Abstract (English)
Visual Place Recognition (VPR) is a critical task in computer vision, traditionally enhanced by re-ranking retrieval results with image matching. However, recent advancements in VPR methods have significantly improved performance, challenging the necessity of re-ranking. In this work, we show that modern retrieval systems often reach a point where re-ranking can degrade results, as current VPR datasets are largely saturated. We propose using image matching as a verification step to assess retrieval confidence, demonstrating that inlier counts can reliably predict when re-ranking is beneficial. Our findings shift the paradigm of retrieval pipelines, offering insights for more robust and adaptive VPR systems. The code is available at https://github.com/FarInHeight/To-Match-or-Not-to-Match.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。