Learning Multi-context Aware Location Representations from Large-scale Geotagged Images

被引:4
|
作者
Yin, Yifang [1 ]
Zhang, Ying [2 ]
Liu, Zhenguang [3 ]
Liang, Yuxuan [1 ]
Wang, Sheng [1 ,4 ]
Shah, Rajiv Ratn [5 ]
Zimmermann, Roger [1 ]
机构
[1] Natl Univ Singapore, Singapore, Singapore
[2] Northwestern Polytech Univ, Xian, Peoples R China
[3] Zhejiang Gongshang Univ, Hangzhou, Peoples R China
[4] Alibaba Grp, Singapore, Singapore
[5] IIIT Delhi, Delhi, India
来源
PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2021 | 2021年
关键词
Location representations; pre-trained neural networks; attentionbased; fusion; geo-aware applications; FEATURES;
D O I
10.1145/3474085.3475268
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
With the ubiquity of sensor-equipped smartphones, it is common to have multimedia documents uploaded to the Internet that have GPS coordinates associated with them. Utilizing such geotags as an additional feature is intuitively appealing for improving the performance of location-aware applications. However, raw GPS coordinates are fine-grained location indicators without any semantic information. Existing methods on geotag semantic encoding mostly extract hand-crafted, application-specific location representations that heavily depend on large-scale supplementary data and thus cannot perform efficiently on mobile devices. In this paper, we present a machine learning based approach, termed GPS2Vec+, which learns rich location representations by capitalizing on the world-wide geotagged images. Once trained, the model has no dependence on the auxiliary data anymore so it encodes geotags highly efficiently by inference. We extract visual and semantic knowledge from image content and user-generated tags, and transfer the information into locations by using geotagged images as a bridge. To adapt to different application domains, we further present an attention-based fusion framework that estimates the importance of the learnt location representations under different contexts for effective feature fusion. Our location representations yield significant performance improvements over the state-of-the-art geotag encoding methods on image classification and venue annotation.
引用
收藏
页码:899 / 907
页数:9
相关论文
共 50 条
  • [21] Learning local equivariant representations for large-scale atomistic dynamics
    Musaelian, Albert
    Batzner, Simon
    Johansson, Anders
    Sun, Lixin
    Owen, Cameron J.
    Kornbluth, Mordechai
    Kozinsky, Boris
    NATURE COMMUNICATIONS, 2023, 14 (01)
  • [22] Learning local equivariant representations for large-scale atomistic dynamics
    Albert Musaelian
    Simon Batzner
    Anders Johansson
    Lixin Sun
    Cameron J. Owen
    Mordechai Kornbluth
    Boris Kozinsky
    Nature Communications, 14
  • [23] Fast Tensor Factorization for Large-Scale Context-Aware Recommendation from Implicit Feedback
    Chou, Szu-Yu
    Jang, Jyh-Shing Roger
    Yang, Yi-Hsuan
    IEEE TRANSACTIONS ON BIG DATA, 2020, 6 (01) : 201 - 208
  • [24] Learning Continuous Word Representations from Large-scale Corpus Through Linear Approach
    Zhang Zhongxia
    Yang Xiaoping
    Ma Qifeng
    Xu Cui
    2015 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC 2015): BIG DATA ANALYTICS FOR HUMAN-CENTRIC SYSTEMS, 2015, : 2678 - 2683
  • [25] MVImgNet: A Large-scale Dataset of Multi-view Images
    Yu, Xianggang
    Xu, Mutian
    Zhang, Yidan
    Liu, Haolin
    Ye, Chongjie
    Wu, Yushuang
    Yan, Zizheng
    Zhu, Chenming
    Xiong, Zhangyang
    Liang, Tianyou
    Chen, Guanying
    Cui, Shuguang
    Han, Xiaoguang
    2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, : 9150 - 9161
  • [26] Context-aware, Composable Anomaly Detection in Large-scale Mobile Networks
    Nguyen Ngoc Nhu Trang
    Hong-Linh Truong
    2023 IEEE 47TH ANNUAL COMPUTERS, SOFTWARE, AND APPLICATIONS CONFERENCE, COMPSAC, 2023, : 183 - 192
  • [27] Distributed cache management for context-aware services in large-scale networks
    Takase, Masaaki
    Sano, Takeshi
    Fukuda, Kenichi
    Chugo, Akira
    MANAGING NEXT GENERATION NETWORKS AND SERVICES, PROCEEDINGS, 2007, 4773 : 31 - +
  • [28] Context Aware Group Nearest Shrunken Centroids in Large-Scale Genomic Studies
    Yang, Juemin
    Han, Fang
    Irizarry, Rafael A.
    Liu, Han
    ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, 33 : 1051 - 1059
  • [29] Context-aware reconfiguration of large-scale surveillance systems: argumentative approach
    Novak, Peter
    Witteveen, Cees
    ARGUMENT & COMPUTATION, 2015, 6 (01) : 3 - 23
  • [30] A Context-aware Service Framework for Large-Scale Ambient Computing Environments
    Satoh, Ichiro
    INTERNATIONAL CONFERENCE ON PERVASIVE SERVICES (ICPS 2009), 2009, : 199 - 207