Unsupervised Domain Adaptation for Depth Prediction from Images

被引:49
|
作者
Tonioni, Alessio [1 ]
Poggi, Matteo [1 ]
Mattoccia, Stefano [1 ]
Di Stefano, Luigi [1 ]
机构
[1] Univ Bologna, Dept Comp Sci & Engn, BO-40126 Bologna, Italy
关键词
Training; Reliability; Estimation; Loss measurement; Computer architecture; Prediction algorithms; Deep learning; depth estimation; unsupervised learning; self-supervised learning; domain adaptation; COST AGGREGATION; STEREO; ACCURATE;
D O I
10.1109/TPAMI.2019.2940948
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
State-of-the-art approaches to infer dense depth measurements from images rely on CNNs trained end-to-end on a vast amount of data. However, these approaches suffer a drastic drop in accuracy when dealing with environments much different in appearance and/or context from those observed at training time. This domain shift issue is usually addressed by fine-tuning on smaller sets of images from the target domain annotated with depth labels. Unfortunately, relying on such supervised labeling is seldom feasible in most practical settings. Therefore, we propose an unsupervised domain adaptation technique which does not require groundtruth labels. Our method relies only on image pairs and leverages on classical stereo algorithms to produce disparity measurements alongside with confidence estimators to assess upon their reliability. We propose to fine-tune both depth-from-stereo as well as depth-from-mono architectures by a novel confidence-guided loss function that handles the measured disparities as noisy labels weighted according to the estimated confidence. Extensive experimental results based on standard datasets and evaluation protocols prove that our technique can address effectively the domain shift issue with both stereo and monocular depth prediction architectures and outperforms other state-of-the-art unsupervised loss functions that may be alternatively deployed to pursue domain adaptation.
引用
收藏
页码:2396 / 2409
页数:14
相关论文
共 50 条
  • [31] Unsupervised Domain Adaptation for Ship Classification via Progressive Feature Alignment: From Optical to SAR Images
    Shi, Yu
    Du, Lan
    Guo, Yuchen
    Du, Yuang
    Li, Yiming
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62
  • [32] Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Training
    Mahmood, Faisal
    Chen, Richard
    Durr, Nicholas J.
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2018, 37 (12) : 2572 - 2581
  • [33] Unsupervised domain adaptation method for segmenting cross-sectional CCA images
    van Knippenberg, Luuk
    van Sloun, Ruud J.G.
    Mischi, Massimo
    de Ruijter, Joerik
    Lopata, Richard
    Bouwman, R. Arthur
    Computer Methods and Programs in Biomedicine, 2022, 225
  • [34] Unsupervised domain adaptation method for segmenting cross-sectional CCA images
    van Knippenberg, Luuk
    van Sloun, Ruud J. G.
    Mischi, Massimo
    de Ruijter, Joerik
    Lopata, Richard
    Bouwman, R. Arthur
    COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2022, 225
  • [35] Unsupervised domain adaptation based fracture segmentation method for core CT images
    Zhao, Xiangxin
    Wang, Xin
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 264
  • [36] A Deep Learning Approach for Unsupervised Domain Adaptation in Multitemporal Remote Sensing Images
    Othman, Essam
    Bazi, Yakoub
    AlHichri, Haikel
    Alajlan, Naif
    2015 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2015, : 2401 - 2404
  • [37] BiFDANet: Unsupervised Bidirectional Domain Adaptation for Semantic Segmentation of Remote Sensing Images
    Cai, Yuxiang
    Yang, Yingchun
    Zheng, Qiyi
    Shen, Zhengwei
    Shang, Yongheng
    Yin, Jianwei
    Shi, Zhongtian
    REMOTE SENSING, 2022, 14 (01)
  • [38] Unsupervised Adversarial Domain Adaptation for Agricultural Land Extraction of Remote Sensing Images
    Zhang, Junbo
    Xu, Shifeng
    Sun, Jun
    Ou, Dinghua
    Wu, Xiaobo
    Wang, Mantao
    REMOTE SENSING, 2022, 14 (24)
  • [39] Adversarial Unsupervised Domain Adaptation for Hand Gesture Recognition Using Thermal Images
    Dayal, Aveen
    Aishwarya, M.
    Abhilash, S.
    Mohan, C. Krishna
    Kumar, Abhinav
    Cenkeramaddi, Linga Reddy
    IEEE SENSORS JOURNAL, 2023, 23 (04) : 3493 - 3504
  • [40] Unsupervised Adversarial Domain Adaptation for Sim-to-Real Transfer of Tactile Images
    Jing, Xingshuo
    Qian, Kun
    Jianu, Tudor
    Luo, Shan
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2023, 72