arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.02431cs.CVcs.LG
收藏

GAMa:跨视图视频地理定位

GAMa: Cross-view Video Geo-localization

Center for Research in Computer Vision, University of Central Florida(中佛罗里达大学计算机视觉研究中心)

Shruti Vyas, Chen Chen, Mubarak Shah

摘要:

现有的跨视图地理定位的工作是基于图像,其中地面全景匹配到航空图像。在这项工作中,我们专注于地面视频,而不是图像,它提供了额外的上下文线索,这是很重要的这项任务。目前还没有针对这个问题的数据集,因此我们提出了GAMa数据集,这是一个包含地面视频和相应航拍图像的大规模数据集。我们还提出了一种新的方法来解决这个问题。在片段级,短视频片段与相应的航拍图像匹配,然后用于获得长视频的视频级地理定位。此外,我们提出了一个层次化的方法,以进一步提高剪辑级的地理定位。这是一个具有挑战性的数据集,未对齐且视野有限,我们提出的方法在1.0英里处实现了19.4%和45.1%的Top-1召回率。代码和数据集可在以下链接获得:https://github.com/svyas23/GAMa。

英文摘要:

The existing work in cross-view geo-localization is based on images where a ground panorama is matched to an aerial image. In this work, we focus on ground videos instead of images which provides additional contextual cues which are important for this task. There are no existing datasets for this problem, therefore we propose GAMa dataset, a large-scale dataset with ground videos and corresponding aerial images. We also propose a novel approach to solve this problem. At clip-level, a short video clip is matched with corresponding aerial image and is later used to get video-level geo-localization of a long video. Moreover, we propose a hierarchical approach to further improve the clip-level geolocalization. It is a challenging dataset, unaligned and limited field of view, and our proposed method achieves a Top-1 recall rate of 19.4% and 45.1% @1.0mile. Code and dataset are available at following link: https://github.com/svyas23/GAMa.

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。

↑