Scene Signatures: Localised and Point-less Features for Localisation
Scene Signatures: Localised and Point-less Features for Localisation
复制标题
DOI：
10.15607/rss.2014.x.023
复制
发表时间：
2014-07
期刊：
影响因子：
--
通讯作者：
C. McManus;B. Upcroft;P. Newman
中科院分区：
文献类型：
--
作者：
C. McManus;B. Upcroft;P. Newman
本文涉及极端光照和天气条件下的定位。我们偏离了传统的基于点特征的方法，因为在剧烈的外观变化下进行匹配是一件脆弱而困难的事情。点特征检测器是固定且严格的程序，它通过图像检查小的低级结构，例如角点或斑点。他们采用相同的标准，适用于所有地方的所有图像。本文采取了相反的观点，并提出了如果我们为每个地方学习一个定制的探测器会发生什么。然后，我们的定位任务变成了管理大量空间索引检测器，我们表明，这在鲁棒性方面产生了极其优越的性能，以换取降低但可以容忍的度量精度。我们提出了一个无监督系统，可以为独特的视觉元素（称为场景签名）生成宽区域检测器，它可以与几乎所有外观变化相关联。我们使用 3 个月内收集的 21 公里数据表明，我们的系统能够从夜间到白天或从夏季到冬季的条件下生成公制定位估计。
This paper is about localising across extreme lighting and weather conditions. We depart from the traditional point-feature-based approach as matching under dramatic appearance changes is a brittle and hard thing. Point feature detectors are fixed and rigid procedures which pass over an image examining small, low-level structure such as corners or blobs. They apply the same criteria applied all images of all places. This paper takes a contrary view and asks what is possible if instead we learn a bespoke detector for every place. Our localisation task then turns into curating a large bank of spatially indexed detectors and we show that this yields vastly superior performance in terms of robustness in exchange for a reduced but tolerable metric precision. We present an unsupervised system that produces broad-region detectors for distinctive visual elements, called scene signatures, which can be associated across almost all appearance changes. We show, using 21km of data collected over a period of 3 months, that our system is capable of producing metric localisation estimates from night-to-day or summer-to-winter conditions.