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
  • [21] Unsupervised Domain Adaptation Based on Progressive Transfer for Ship Detection: From Optical to SAR Images
    Shi, Yu
    Du, Lan
    Guo, Yuchen
    Du, Yuang
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
  • [22] Unsupervised Adversarial Visual Level Domain Adaptation for Learning Video Object Detectors from Images
    Lahiri, Avisek
    Ragireddy, Charan
    Biswas, Prabir Kumar
    Mitra, Pabitra
    2019 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV), 2019, : 1807 - 1815
  • [23] Unsupervised Domain Adaptation for Learning Eye Gaze from a Million Synthetic Images: An Adversarial Approach
    Lahiri, Avisek
    Agarwalla, Abhinav
    Biswas, Prabir Kumar
    ELEVENTH INDIAN CONFERENCE ON COMPUTER VISION, GRAPHICS AND IMAGE PROCESSING (ICVGIP 2018), 2018,
  • [24] Unsupervised Personal Thermal Comfort Prediction via Adversarial Domain Adaptation
    Das, Hari Prasanna
    Schiavon, Stefano
    Spanos, Costas J.
    BUILDSYS'21: PROCEEDINGS OF THE 2021 ACM INTERNATIONAL CONFERENCE ON SYSTEMS FOR ENERGY-EFFICIENT BUILT ENVIRONMENTS, 2021, : 230 - 231
  • [25] UDA-DP: Unsupervised Domain Adaptation for Software Defect Prediction
    Huang, Xiaosong
    Wu, Yifan
    Liu, Hongyi
    Li, Ying
    Yu, Hao
    Guo, Dadi
    Wu, Zhonghai
    2023 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE ANALYSIS, EVOLUTION AND REENGINEERING, SANER, 2023, : 308 - 318
  • [26] Transfer learning with unsupervised domain adaptation for personal thermal comfort prediction
    Yang, Chuangkang
    Taniguchi, Keiichiro
    Akashi, Yasunori
    ENERGY AND BUILDINGS, 2025, 332
  • [27] Robust Unsupervised Domain Adaptation from A Corrupted Source
    Yu, Shuyang
    Zhu, Zhuangdi
    Liu, Boyang
    Jain, Anil K.
    Zhou, Jiayu
    2022 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), 2022, : 1299 - 1304
  • [28] Unsupervised Domain Adaptation by Backpropagation
    Ganin, Yaroslav
    Lempitsky, Victor
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 37, 2015, 37 : 1180 - 1189
  • [29] Unsupervised variational domain adaptation
    Li, Yundong
    Ge, Yizheng
    Lin, Chen
    Wang, Guan
    MACHINE LEARNING, 2025, 114 (03)
  • [30] Domain Adaptation for Unsupervised Cancer Detection: An Application for Skin Whole Slides Images from an Interhospital Dataset
    Garcia-de-la-Puente, Natalia P.
    Lopez-Perez, Miguel
    Launet, Laetitia
    Naranjo, Valery
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2024, PT IV, 2024, 15004 : 58 - 68