Image Colorization Using the Global Scene-Context Style and Pixel-Wise Semantic Segmentation

被引:6
|
作者
Tram-Tran Nguyen-Quynh [1 ]
Kim, Soo-Hyung [1 ]
Nhu-Tai Do [1 ]
机构
[1] Chonnam Natl Univ, Dept Artificial Intelligence Convergence, Gwangju 61186, South Korea
基金
新加坡国家研究基金会;
关键词
Image color analysis; Semantics; Image segmentation; Computer architecture; Gray-scale; Task analysis; Encoding; Image colorization; soft-encoding; u-net; scene-context classification; semantic segmentation; COLOR;
D O I
10.1109/ACCESS.2020.3040737
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we present an encoder-decoder architecture that exploits global and local semantics for the automatic image colorization problem. For the global semantics, the low-level encoding features are fine-tuned by the scene-context classification to integrate the global image style. Moreover, the architecture deals with the uncertainty and relations among the scene styles based on the label smoothing and pre-trained weights from Places365. For local semantics, three branches learn the mutual benefits at the pixel-level, in which average and multi-modal distributions are respectively created from regression and soft-encoding branches, while the segmentation branch determines to which object the pixel belongs. Our experiments, which involve training with the Coco-Stuff dataset and validation on DIV2K, Places365, and ImageNet, show that our results are very encouraging.
引用
收藏
页码:214098 / 214114
页数:17
相关论文
共 50 条
  • [41] A pixel-wise lab elle d dataset of Moroccan aircraft emergency landing sites for semantic segmentation applications
    Illi, Adil
    Bouzaachane, Khadija
    El Hadaj, Salah
    El Guarmah, El Mahdi
    DATA IN BRIEF, 2024, 54
  • [42] Efficient Pixel-Wise SVD Required for Image Processing Using the Color Line Feature
    Shirai, Keiichiro
    Ito, Yuya
    Miyao, Hidetoshi
    Maruyama, Minoru
    IEEE ACCESS, 2021, 9 : 79449 - 79460
  • [43] Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function
    Strohmann, Tobias
    Bugelnig, Katrin
    Breitbarth, Eric
    Wilde, Fabian
    Steffens, Thomas
    German, Holger
    Requena, Guillermo
    SCIENTIFIC REPORTS, 2019, 9 (1)
  • [44] Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function
    Tobias Strohmann
    Katrin Bugelnig
    Eric Breitbarth
    Fabian Wilde
    Thomas Steffens
    Holger Germann
    Guillermo Requena
    Scientific Reports, 9
  • [45] High-Speed Imaging using CMOS Image Sensor with Quasi Pixel-Wise Exposure
    Sonoda, T.
    Nagahara, H.
    Endo, K.
    Sugiyama, Y.
    Taniguchi, R.
    SELECTED PAPERS FROM THE 31ST INTERNATIONAL CONGRESS ON HIGH-SPEED IMAGING AND PHOTONICS, 2017, 10328
  • [46] BEMD BASED ULTRASOUND IMAGE SPECKLE REDUCTION TECHNIQUE USING PIXEL-WISE WIENER FILTERING
    Gupta, Bhawna
    Khandelwal, Vineet
    ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING, 2021, 19 (02) : 168 - 178
  • [47] High-Speed Imaging Using CMOS Image Sensor With Quasi Pixel-Wise Exposure
    Yoshida, Michitaka
    Sonoda, Toshiki
    Nagahara, Hajime
    Endo, Kenta
    Sugiyama, Yukinobu
    Taniguchi, Rin-ichiro
    IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING, 2020, 6 (06) : 463 - 476
  • [48] Lightweight pixel-wise segmentation for efficient concrete crack detection using hierarchical convolutional neural network
    Kim, Jin
    Shim, Seungbo
    Cha, Yohan
    Cho, Gye-Chun
    SMART MATERIALS AND STRUCTURES, 2021, 30 (04)
  • [49] PCT-Net: Full Resolution Image Harmonization Using Pixel-Wise Color Transformations
    Guerreiro, Julian Jorge Andrade
    Nakazawa, Mitsuru
    Stenger, Bjorn
    2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR, 2023, : 5917 - 5926
  • [50] High-Speed Imaging using CMOS Image Sensor with Quasi Pixel-Wise Exposure
    Sonoda, Toshiki
    Nagahara, Hajime
    Endo, Kenta
    Sugiyama, Yukinobu
    Taniguchi, Rin-ichiro
    2016 IEEE INTERNATIONAL CONFERENCE ON COMPUTATIONAL PHOTOGRAPHY (ICCP), 2016, : 81 - 91