arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2204.03444cs.CV
收藏

Deep Visual Geo-localization Benchmark(深度视觉地理定位基准)

Deep Visual Geo-localization Benchmark

Politecnico di Torino(都灵理工大学);CINI(意大利国家信息与大学计算联盟);NAVER LABS Europe(NAVER欧洲实验室)

Gabriele Berton, Riccardo Mereu, Gabriele Trivigno, Carlo Masone, Gabriela Csurka, Torsten Sattler, Barbara Caputo

AI总结:

该研究提出一个开源深度视觉地理定位基准框架,可灵活调整定位流水线组件,通过大量实验给出组件选择依据,并验证简单工程技巧可在保证性能的同时降低时间与存储成本。

摘要:

本文提出了一种面向视觉地理定位(Visual Geo-localization,VG)的新型开源基准测试框架,可用于构建、训练和测试多种常用架构,且能灵活调整地理定位流水线的各个组件。该框架的目标分为两方面:其一,探究VG流水线中不同组件与设计选择如何从性能(recall@N指标)和系统需求(如执行时间、内存消耗)两方面影响最终结果;其二,建立一套用于对比不同方法的系统性评估协议。借助所提框架,我们开展了大量实验,为根据使用场景与需求选择骨干网络、聚合方法及负样本挖掘策略提供了依据。我们还评估了预处理/后处理、数据增强、图像缩放等工程技术的影响,结果表明通过一些相对简单的操作就能提升性能:例如将图像分辨率降至80%,可在取得相近结果的同时,减少36%的特征提取时间与数据集存储需求。代码与训练好的模型可访问https://deep-vg-bench.herokuapp.com/获取。

英文摘要:

In this paper, we propose a new open-source benchmarking framework for Visual Geo-localization (VG) that allows to build, train, and test a wide range of commonly used architectures, with the flexibility to change individual components of a geo-localization pipeline. The purpose of this framework is twofold: i) gaining insights into how different components and design choices in a VG pipeline impact the final results, both in terms of performance (recall@N metric) and system requirements (such as execution time and memory consumption); ii) establish a systematic evaluation protocol for comparing different methods. Using the proposed framework, we perform a large suite of experiments which provide criteria for choosing backbone, aggregation and negative mining depending on the use-case and requirements. We also assess the impact of engineering techniques like pre/post-processing, data augmentation and image resizing, showing that better performance can be obtained through somewhat simple procedures: for example, downscaling the images' resolution to 80% can lead to similar results with a 36% savings in extraction time and dataset storage requirement. Code and trained models are available at https://deep-vg-bench.herokuapp.com/.

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。

↑