Deep Learning Approaches Assessment for Underwater Scene Understanding and Egomotion Estimation

被引:0
作者
Teixeira, Bernardo [1 ,3 ]
Silva, Hugo [1 ]
Matos, Anibal [1 ,3 ]
Silva, Eduardo [1 ,2 ]
机构
[1] Porto Polytech Inst, INESC IBC Inst Syst & Comp Engn Technol & Sci, Porto, Portugal
[2] Porto Polytech Inst, ISEP Sch Engn, Porto, Portugal
[3] Univ Porto, FEUP Fac Engn, Porto, Portugal
来源
OCEANS 2019 MTS/IEEE SEATTLE | 2019年
关键词
Artificial intelligence; Computer vision; Deep learning; Visual Odometry; Robot navigation;
D O I
10.23919/oceans40490.2019.8962872
中图分类号
U6 [水路运输]; P75 [海洋工程];
学科分类号
0814 ; 081505 ; 0824 ; 082401 ;
摘要
This paper address the use of deep learning approaches for visual based navigation in confined underwater environments. State-of-the-art algorithms have shown the tremendous potential deep learning architectures can have for visual navigation implementations, though they are still mostly outperformed by classical feature-based techniques. In this work, we apply current state-of-the-art deep learning methods for visual-based robot navigation to the more challenging underwater environment, providing both an underwater visual dataset acquired in real operational mission scenarios and an assessment of state-of-the-art algorithms on the underwater context. We extend current work by proposing a novel pose optimization architecture for the purpose of correcting visual odometry estimate drift using a Visual-Inertial fusion network, consisted of a neural network architecture anchored on an Inertial supervision learning scheme. Our Visual-Inertial Fusion Network was shown to improve results an average of 50% for trajectory estimates, also producing more visually consistent trajectory estimates for both our underwater application scenarios.
引用
收藏
页数:9
相关论文
共 50 条
  • [21] Indoor Scene Classification through Dual-Stream Deep Learning: A Framework for Improved Scene Understanding in Robotics
    Khan, Sultan Daud
    Othman, Kamal M.
    COMPUTERS, 2024, 13 (05)
  • [22] Understanding gut microbiome-based machine learning platforms: A review on therapeutic approaches using deep learning
    Malakar, Shilpa
    Sutaoney, Priya
    Madhyastha, Harishkumar
    Shah, Kamal
    Chauhan, Nagendra Singh
    Banerjee, Paromita
    CHEMICAL BIOLOGY & DRUG DESIGN, 2024, 103 (03)
  • [23] StepNet-Deep Learning Approaches for Step Length Estimation
    Klein, Itzik
    Asraf, Omri
    IEEE ACCESS, 2020, 8 : 85706 - 85713
  • [24] Deep semantic segmentation for visual scene understanding of soil types
    Zamani, Vahid
    Taghaddos, Hosein
    Gholipour, Yaghob
    Pourreza, Hamidreza
    AUTOMATION IN CONSTRUCTION, 2022, 140
  • [25] Machine learning and deep learning approaches in IoT
    Javed A.
    Awais M.
    Shoaib M.
    Khurshid K.S.
    Othman M.
    PeerJ Computer Science, 2023, 9
  • [26] A Survey on Deep Learning-Driven Remote Sensing Image Scene Understanding: Scene Classification, Scene Retrieval and Scene-Guided Object Detection
    Gu, Yating
    Wang, Yantian
    Li, Yansheng
    APPLIED SCIENCES-BASEL, 2019, 9 (10):
  • [27] Lung cancer survival period prediction and understanding: Deep learning approaches
    Doppalapudi, Shreyesh
    Qiu, Robin G.
    Badr, Youakim
    INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS, 2021, 148
  • [28] Deep Learning Aided Channel Estimation and Signal Detection for Underwater Optical Communication
    Jia S.
    Aiping H.
    Linwei T.
    Zhongguo Jiguang/Chinese Journal of Lasers, 2022, 49 (17):
  • [29] Deep Learning for age Estimation
    Ammous, Donia
    Kammoun, Fahmi
    Masmoudi, Nouri
    2024 IEEE 7TH INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES, SIGNAL AND IMAGE PROCESSING, ATSIP 2024, 2024, : 322 - 329
  • [30] Learning Direct Optimization for scene understanding
    Romaszko, Lukasz
    Williams, Christopher K., I
    Winn, John
    PATTERN RECOGNITION, 2020, 105