Joint Deep Estimation of Intrinsic and Dichromatic Image Decomposition

被引:0
|
作者
Ha, Jeong-Won [1 ]
Lee, Kang-Kyu [1 ]
Kim, Jong-Ok [1 ]
机构
[1] Korea Univ, Sch Elect Engn, Seoul 02841, South Korea
基金
新加坡国家研究基金会;
关键词
Reflectivity; Reflection; Lighting; Image decomposition; Image color analysis; Estimation; Surface treatment; Intrinsic image decomposition; dichromatic model; color constancy; AC light; high-speed video; HIGHLIGHT REMOVAL; REFLECTION; SEPARATION; COLOR;
D O I
10.1109/ACCESS.2023.3271114
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper proposes an image formation model that jointly combines dichromatic and intrinsic image decomposition models. The two decomposition models analyze image formation process from a different perspective, and they can be combined synergistically. It is confirmed that the proposed method performs better than the individual decomposition. The joint estimation and the study of the decomposition order ('intrinsic + dichromatic' or 'dichromatic + intrinsic') are the first attempt to the best of our knowledge. It was confirmed that the proposed 'intrinsic + dichromatic' is more optimal through experimental evaluations. We also exploit the temporal property of AC light sources, which can further improve the decomposition performance. The experimental results show that the proposed model can make an accurate image decomposition and achieve a remarkable color constancy performance.
引用
收藏
页码:41770 / 41782
页数:13
相关论文
共 50 条
  • [11] Deep intrinsic image decomposition under colored AC light sources
    Lee, Kang-Kyu
    Ha, Jeong-Won
    Yoo, Jun-Sang
    Kim, Jong-Ok
    MULTIMEDIA TOOLS AND APPLICATIONS, 2024, 83 (05) : 14775 - 14795
  • [12] Deep intrinsic image decomposition under colored AC light sources
    Kang-Kyu Lee
    Jeong-Won Ha
    Jun-Sang Yoo
    Jong-Ok Kim
    Multimedia Tools and Applications, 2024, 83 : 14775 - 14795
  • [13] Learning Intrinsic Image Decomposition by Deep Neural Network with Perceptual Loss
    Han, Guangyun
    Xie, Xiaohua
    Zheng, Wei-Shi
    Lai, Jianhuang
    2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2018, : 91 - 96
  • [14] Joint Deep Learning for Covariate Decomposition and Treatment Effect Estimation
    Shen, Hanlei
    Zhang, Hu
    Song, Xinyuan
    STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 2023, 30 (04) : 547 - 559
  • [15] Single Image Intrinsic Decomposition Without a Single Intrinsic Image
    Ma, Wei-Chiu
    Chu, Hang
    Zhou, Bolei
    Urtasun, Raquel
    Torralba, Antonio
    COMPUTER VISION - ECCV 2018, PT XIV, 2018, 11218 : 211 - 229
  • [16] A Review of Intrinsic Image Decomposition
    Liu, Siyuan
    Jiang, Xiaoyue
    Liu, Letian
    Xia, Zhaoqiang
    Dang, Sihang
    Feng, Xiaoyi
    2024 3RD INTERNATIONAL CONFERENCE ON IMAGE PROCESSING AND MEDIA COMPUTING, ICIPMC 2024, 2024, : 254 - 261
  • [17] Deep Dichromatic Model Estimation Under AC Light Sources
    Yoo, Jun-Sang
    Lee, Chan-Ho
    Kim, Jong-Ok
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2021, 30 : 7064 - 7073
  • [18] Unsupervised Deep Single-Image Intrinsic Decomposition using Illumination-Varying Image Sequences
    Lettry, L.
    Vanhoey, K.
    Van Gool, L.
    COMPUTER GRAPHICS FORUM, 2018, 37 (07) : 409 - 419
  • [19] Unified Depth Prediction and Intrinsic Image Decomposition from a Single Image via Joint Convolutional Neural Fields
    Kim, Seungryong
    Park, Kihong
    Sohn, Kwanghoon
    Lin, Stephen
    COMPUTER VISION - ECCV 2016, PT VIII, 2016, 9912 : 143 - 159
  • [20] Classification of Hyperspectral Images based on Intrinsic Image Decomposition and Deep Convolutional Neural Network
    Beirami, Behnam Asghari
    Mokhtarzade, Mehdi
    2020 6TH IRANIAN CONFERENCE ON SIGNAL PROCESSING AND INTELLIGENT SYSTEMS (ICSPIS), 2020,