Exploring ubiquitous relations for boosting classification and localization

被引：0
|
作者：
Sun, Xin [1 ]
Chen, Changrui [1 ]
Dong, Junyu [1 ]
Liu, Dan [1 ]
Hu, Guosheng [2 ]
机构：
[1] Department of Computer Science and Technology, Ocean University of China, China
[2] School of Electronics Electrical Engineering and Computer Science, Queen's University Belfast, United Kingdom
来源：
Knowledge-Based Systems | 2021年 / 196卷
关键词：
Although the weakly supervised learning can effectively avoid the tedious data annotating process of deep learning approaches; the performance is still in urgent need of enhancement. In this paper; we endeavor to mine a ubiquitous and fundamental knowledge—Relation; to boost several existing classification and localization models without changing the original structure. We first propose a universal relation exploring scheme to mine the relations among entities. This scheme can be specialized into different instantiations including object; superpixel and pixel relations to stimulating different learning models. We adopt the object relations on a few-shot classification model to concentrate on the dominant object; and to boost its discriminative capacity. The superpixel relation is utilized to improve the performance of the saliency object detection models. The sensitivity of the pixel relations to the uncertain regions makes it suitable for distinguishing the disputed area in saliency detection results. Our experiments demonstrate that all the three relation instantiations can significantly boost the performance of the state-of-the-art learning models and optimize the visual result. © 2020 Elsevier B.V;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
引用
收藏
相关论文
共 50 条
  • [1] Exploring ubiquitous relations for boosting classification and localization
    Sun, Xin
    Chen, Changrui
    Dong, Junyu
    Liu, Dan
    Hu, Guosheng
    KNOWLEDGE-BASED SYSTEMS, 2020, 196
  • [2] BOOSTING GLOBAL SCENE CLASSIFICATION ACCURACY BY DISCRIMINATIVE REGION LOCALIZATION
    Thanh Duc Ngo
    Duy-Dinh Le
    Satoh, Shin'ichi
    2011 18TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2011, : 1053 - 1056
  • [3] Exploring Relations of Visual Codes for Image Classification
    Huang, Yongzhen
    Huang, Kaiqi
    Wang, Chong
    Tan, Tieniu
    2011 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2011, : 1649 - 1656
  • [4] Boosting Dependable Ubiquitous Computing: A Case Study
    Fernandez, C.
    Cortinas, R.
    Larrea, M.
    IEEE LATIN AMERICA TRANSACTIONS, 2014, 12 (03) : 442 - 448
  • [5] Classification - the ubiquitous challenge
    Steinley, Douglas
    JOURNAL OF CLASSIFICATION, 2007, 24 (01) : 144 - 152
  • [6] Boosting Powerline Communications for Ubiquitous Connectivity in Enterprises
    Ali, Kamran
    Pefkianakis, Ioaimis
    Liu, Alex X.
    Kim, Kyu-Han
    2016 IEEE 24TH INTERNATIONAL CONFERENCE ON NETWORK PROTOCOLS (ICNP), 2016,
  • [7] Hybrid Localization for Ubiquitous Services
    Ficco, Massimo
    Aversa, Rocco
    2013 IEEE 10TH INTERNATIONAL CONFERENCE ON AND 10TH INTERNATIONAL CONFERENCE ON AUTONOMIC AND TRUSTED COMPUTING (UIC/ATC) UBIQUITOUS INTELLIGENCE AND COMPUTING, 2013, : 1 - 8
  • [8] Exploring Inter-Node Relations in CNNs for Environmental Sound Classification
    Song, Hongwei
    Deng, Shiwen
    Han, Jiqing
    IEEE SIGNAL PROCESSING LETTERS, 2022, 29 : 154 - 158
  • [9] Resealed Boosting in Classification
    Wang, Yao
    Liao, Xu
    Li, Shaobo
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (09) : 2598 - 2610
  • [10] Everywhere: A Framework for Ubiquitous Indoor Localization
    Mansour, Ahmed
    Ye, Junhua
    Li, Yaxin
    Luo, Huan
    Wang, Jingxian
    Weng, Duojie
    Chen, Wu
    IEEE INTERNET OF THINGS JOURNAL, 2023, 10 (06) : 5095 - 5113
12345