Semi-Supervised Semantic Image Segmentation by Deep Diffusion Models and Generative Adversarial Networks

被引:0
|
作者
Diaz-Frances, Jose Angel [1 ]
Fernandez-Rodriguez, Jose David [1 ]
Thurnhofer-Hemsi, Karl [1 ]
Lopez-Rubio, Ezequiel [1 ]
机构
[1] Univ Malaga, ITIS Software, Calle Arquitecto Francisco Penalosa 18, Malaga 29010, Spain
关键词
Semantic segmentation; semi-supervised; diffusion model;
D O I
10.1142/S0129065724500576
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Typically, deep learning models for image segmentation tasks are trained using large datasets of images annotated at the pixel level, which can be expensive and highly time-consuming. A way to reduce the amount of annotated images required for training is to adopt a semi-supervised approach. In this regard, generative deep learning models, concretely Generative Adversarial Networks (GANs), have been adapted to semi-supervised training of segmentation tasks. This work proposes MaskGDM, a deep learning architecture combining some ideas from EditGAN, a GAN that jointly models images and their segmentations, together with a generative diffusion model. With careful integration, we find that using a generative diffusion model can improve EditGAN performance results in multiple segmentation datasets, both multi-class and with binary labels. According to the quantitative results obtained, the proposed model improves multi-class image segmentation when compared to the EditGAN and DatasetGAN models, respectively, by 4.5% and 5.0%. Moreover, using the ISIC dataset, our proposal improves the results from other models by up to 11% for the binary image segmentation approach.
引用
下载
收藏
页数:16
相关论文
共 50 条
  • [1] SEMI-SUPERVISED SEMANTIC GENERATIVE NETWORKS FOR REMOTE SENSING IMAGE SEGMENTATION
    Lu, Wanxuan
    Jin, Jidong
    Sun, Xian
    Fu, Kun
    IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 6386 - 6389
  • [2] Semi-supervised semantic segmentation based on Generative Adversarial Networks for remote sensing images
    Liu Yu-Xi
    Zhang Bo
    Wang Bin
    JOURNAL OF INFRARED AND MILLIMETER WAVES, 2020, 39 (04) : 473 - 482
  • [3] GENERATIVE ADVERSARIAL SEMI-SUPERVISED NETWORK FOR MEDICAL IMAGE SEGMENTATION
    Li, Chuchen
    Liu, Huafeng
    2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI), 2021, : 303 - 306
  • [4] Medical image segmentation with generative adversarial semi-supervised network
    Li, Chuchen
    Liu, Huafeng
    PHYSICS IN MEDICINE AND BIOLOGY, 2021, 66 (24):
  • [5] Semi-Supervised Semantic Image Segmentation using Dual Discriminator Adversarial Networks
    Liu, Beibei
    Hua, Bei
    ELEVENTH INTERNATIONAL CONFERENCE ON DIGITAL IMAGE PROCESSING (ICDIP 2019), 2019, 11179
  • [6] Semi-supervised semantic segmentation using an improved generative adversarial network
    Xu, Di
    Wang, Zhili
    JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2021, 40 (05) : 9709 - 9719
  • [7] A semi-supervised image segmentation method based on generative adversarial network
    Nie, Wei
    Gou, Peng
    Liu, Yang
    Zhou, Tianyu
    Xu, Nuo
    Wang, Peng
    Du, QiQi
    IEEE Joint International Information Technology and Artificial Intelligence Conference (ITAIC), 2022, 2022-June : 1217 - 1223
  • [8] A Semi-supervised Encoder Generative Adversarial Networks Model for Image Classification
    Fu, Xiao
    Shen, Yuan-Tong
    Li, Hong-Wei
    Cheng, Xiao-Mei
    Zidonghua Xuebao/Acta Automatica Sinica, 2020, 46 (03): : 531 - 539
  • [9] Semi-supervised image attribute editing using generative adversarial networks
    Dogan, Yahya
    Keles, Hacer Yalim
    NEUROCOMPUTING, 2020, 401 (401) : 338 - 352
  • [10] SEMI-SUPERVISED VARIATIONAL GENERATIVE ADVERSARIAL NETWORKS FOR HYPERSPECTRAL IMAGE CLASSIFICATION
    Wang, Hao
    Tao, Chao
    Qi, Ji
    Li, HaiFeng
    Tang, YuQi
    2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019), 2019, : 9792 - 9794