期刊
IEEE TRANSACTIONS ON IMAGE PROCESSING
卷 31, 期 -, 页码 3780-3792出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2022.3175601
关键词
Feature extraction; Convolution; Task analysis; Location awareness; Visualization; Visual systems; Representation learning; Geo-localization; representation learning; keypoint; attention
资金
- National Natural Science Foundation of China [61876159, 61806172]
- European Union [951911]
- Progetti di Ricerca di Interesse Nazionale Project CREATIVE Prot [2020ZSL9F9]
- Zhejiang Lab's International Talent Fund for Young Professionals [ZJ2020GZ021]
This paper studies the cross-view geo-localization problem and proposes a framework called RK-Net to jointly learn discriminative representation and detect salient keypoints using the USAM module. The integration of USAM enables end-to-end joint learning, simplifies implementation, and enhances overall performance, achieving competitive accuracy on challenging datasets.
In this paper, we study the cross-view geo-localization problem to match images from different viewpoints. The key motivation underpinning this task is to learn a discriminative viewpoint-invariant visual representation. Inspired by the human visual system for mining local patterns, we propose a new framework called RK-Net to jointly learn the discriminative Representation and detect salient Keypoints with a single Network. Specifically, we introduce a Unit Subtraction Attention Module (USAM) that can automatically discover representative keypoints from feature maps and draw attention to the salient regions. USAM contains very few learning parameters but yields significant performance improvement and can be easily plugged into different networks. We demonstrate through extensive experiments that (1) by incorporating USAM, RK-Net facilitates end-to-end joint learning without the prerequisite of extra annotations. Representation learning and keypoint detection are two highly-related tasks. Representation learning aids keypoint detection. Keypoint detection, in turn, enriches the model capability against large appearance changes caused by viewpoint variants. (2) USAM is easy to implement and can be integrated with existing methods, further improving the state-of-the-art performance. We achieve competitive geo-localization accuracy on three challenging datasets, i. e., University-1652, CVUSA and CVACT. Our code is available at https://github.com/AggMan96/RK-Net.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据