Adversarial Domain Feature Adaptation for Bronchoscopic Depth Estimation

被引:14
|
作者
Karaoglu, Mert Asim [1 ,2 ]
Brasch, Nikolas [2 ]
Stollenga, Marijn [1 ]
Wein, Wolfgang [1 ]
Navab, Nassir [2 ,3 ]
Tombari, Federico [2 ,4 ]
Ladikos, Alexander [1 ]
机构
[1] ImFus GmbH, Munich, Germany
[2] Tech Univ Munich, Comp Aided Med Procedures, Munich, Germany
[3] Johns Hopkins Univ, Comp Aided Med Procedures, Baltimore, MD USA
[4] Google, Zurich, Switzerland
来源
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2021, PT IV | 2021年 / 12904卷
关键词
Bronchoscopy; Depth estimation; Domain adaptation; RECONSTRUCTION; NAVIGATION; CT;
D O I
10.1007/978-3-030-87202-1_29
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Depth estimation from monocular images is an important task in localization and 3D reconstruction pipelines for bronchoscopic navigation. Various supervised and self-supervised deep learning-based approaches have proven themselves on this task for natural images. However, the lack of labeled data and the bronchial tissue's feature-scarce texture make the utilization of these methods ineffective on bronchoscopic scenes. In this work, we propose an alternative domain-adaptive approach. Our novel two-step structure first trains a depth estimation network with labeled synthetic images in a supervised manner; then adopts an unsupervised adversarial domain feature adaptation scheme to improve the performance on real images. The results of our experiments show that the proposed method improves the network's performance on real images by a considerable margin and can be employed in 3D reconstruction pipelines.
引用
收藏
页码:300 / 310
页数:11
相关论文
共 50 条
  • [41] Adversarial joint domain adaptation of asymmetric feature mapping based on least squares distance
    Yuan, Yumeng
    Li, Yuhua
    Zhu, Zhenlong
    Li, Ruixuan
    Gu, Xiwu
    PATTERN RECOGNITION LETTERS, 2020, 136 : 251 - 256
  • [42] Multi-view Feature Learning for the Over-penalty in Adversarial Domain Adaptation
    Zhang, Yuhong
    Wu, Jianqing
    Zhang, Qi
    Hu, Xuegang
    DATA INTELLIGENCE, 2024, 6 (01) : 183 - 200
  • [43] Multi-view Feature Learning for the Over-penalty in Adversarial Domain Adaptation
    Yuhong Zhang
    Jianqing Wu
    Qi Zhang
    Xuegang Hu
    Data Intelligence, 2024, 6 (01) : 183 - 200
  • [44] Backprop Induced Feature Weighting for Adversarial Domain Adaptation with Iterative Label Distribution Alignment
    Westfechtel, Thomas
    Yeh, Hao-Wei
    Meng, Qier
    Mukuta, Yusuke
    Harada, Tatsuya
    2023 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV), 2023, : 392 - 401
  • [45] Privacy-Preserving Federated Adversarial Domain Adaptation Over Feature Groups for Interpretability
    Kang, Yan
    He, Yuanqin
    Luo, Jiahuan
    Fan, Tao
    Liu, Yang
    Yang, Qiang
    IEEE TRANSACTIONS ON BIG DATA, 2024, 10 (06) : 879 - 890
  • [46] Sleep Quality Estimation with Adversarial Domain Adaptation: From Laboratory to Real Scenario
    Tong, Jia-Jun
    Luo, Yun
    Ma, Bo-Qun
    Zheng, Wei-Long
    Lu, Bao-Liang
    Song, Xiao-Qi
    Ma, Shi-Wei
    2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2018,
  • [47] Unsupervised adversarial domain adaptation leaf counting with Bayesian loss density estimation
    Jie Mei
    Kaiqiong Sun
    Signal, Image and Video Processing, 2023, 17 : 1503 - 1509
  • [48] Unsupervised adversarial domain adaptation leaf counting with Bayesian loss density estimation
    Mei, Jie
    Sun, Kaiqiong
    SIGNAL IMAGE AND VIDEO PROCESSING, 2023, 17 (04) : 1503 - 1509
  • [49] Geometry-Aware Symmetric Domain Adaptation for Monocular Depth Estimation
    Zhao, Shanshan
    Fu, Huan
    Gong, Mingming
    Tao, Dacheng
    2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 9780 - 9790
  • [50] Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation
    Zhang, Jiehua
    Li, Liang
    Yan, Chenggang
    Wang, Zhan
    Xu, Changliang
    Zhang, Jiyong
    Chen, Chuqiao
    ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS, 2024, 20 (09)