Multi-scale Information Assembly for Image Matting

被引:10
|
作者
Qiao, Yu [1 ]
Liu, Yuhao [1 ]
Zhu, Qiang [1 ]
Yang, Xin [1 ]
Wang, Yuxin [1 ]
Zhang, Qiang [1 ]
Wei, Xiaopeng [1 ]
机构
[1] Dalian Univ Technol, Coll Comp Sci, Dalian, Peoples R China
基金
中国国家自然科学基金;
关键词
CCS Concepts; Image representations; • Computing methodologies $Ar Image segmentation;
D O I
10.1111/cgf.14168
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Image matting is a long-standing problem in computer graphics and vision, mostly identified as the accurate estimation of the foreground in input images. We argue that the foreground objects can be represented by different-level information, including the central bodies, large-grained boundaries, refined details, etc. Based on this observation, in this paper, we propose a multi-scale information assembly framework (MSIA-matte) to pull out high-quality alpha mattes from single RGB images. Technically speaking, given an input image, we extract advanced semantics as our subject content and retain initial CNN features to encode different-level foreground expression, then combine them by our well-designed information assembly strategy. Extensive experiments can prove the effectiveness of the proposed MSIA-matte, and we can achieve state-of-the-art performance compared to most existing matting networks.
引用
收藏
页码:565 / 574
页数:10
相关论文
共 50 条
  • [31] Multi-Focus Image Fusion Method Based on Multi-Scale Decomposition of Information Complementary
    Wan, Hui
    Tang, Xianlun
    Zhu, Zhiqin
    Li, Weisheng
    ENTROPY, 2021, 23 (10)
  • [32] Image Matting with Color and Depth Information
    Lu, Ting
    Li, Shutao
    2012 21ST INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR 2012), 2012, : 3787 - 3790
  • [33] INFORMATION GRANULES IN MULTI-SCALE ORDERED INFORMATION SYSTEMS
    Wu, Wei-Zhi
    Gu, Shen-Ming
    Wang, Xia
    PROCEEDINGS OF 2015 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS (ICMLC), VOL. 1, 2015, : 182 - 187
  • [34] Medical image super-resolution reconstruction via multi-scale information distillation network under multi-scale geometric transform domain
    Wang H.
    Sun T.
    Shengwu Yixue Gongchengxue Zazhi/Journal of Biomedical Engineering, 2022, 39 (05): : 887 - 896
  • [35] Multi-Scale Region-Aware Implicit Neural Network for Medical Images Matting
    Xu, Yanyu
    Xia, Yingzhi
    Fu, Huazhu
    Goh, Rick Siow Mong
    Liu, Yong
    Xu, Xinxing
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2024, PT IX, 2024, 15009 : 467 - 477
  • [36] Adaptive Multi-scale Information Fusion Based on Dynamic Receptive Field for Image-to-image Translation
    Yin Mengxiao
    Lin Zhenfeng
    Yang Feng
    JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY, 2021, 43 (08) : 2386 - 2394
  • [37] Remote sensing image cloud removal based on multi-scale spatial information perception
    Dou, Aozhe
    Hao, Yang
    Liu, Weifeng
    Li, Liangliang
    Wang, Zhenzhong
    Liu, Baodi
    MULTIMEDIA SYSTEMS, 2024, 30 (05)
  • [38] Multi-Scale Factor Image Super-Resolution Algorithm with Information Distillation Network
    Cheng, Yu
    Chen, Shuai
    Liao, Zeyu
    Zhou, Niujun
    APPLIED SCIENCES-BASEL, 2022, 12 (09):
  • [39] Hyperspectral Image Classification Based on Spectral and Spatial Information Using Multi-Scale ResNet
    Wang, Zong-Yue
    Xia, Qi-Ming
    Yan, Jing-Wen
    Xuan, Shu-Qi
    Su, Jin-He
    Yang, Cheng-Fu
    APPLIED SCIENCES-BASEL, 2019, 9 (22):
  • [40] Multi-Scale Information Can Do More: Attention Aggregation Mechanism for Image Compression
    Li, Bo
    Li, Yongjun
    Li, Yao
    Luo, Jincheng
    Li, Chaoyue
    2023 DATA COMPRESSION CONFERENCE, DCC, 2023, : 348 - 348