Replay-Based Online Adaptation for Unsupervised Deep Visual Odometry

被引：0
|
作者：
Kuznietsov, Yevhen [1 ]
Proesmans, Marc [1 ]
Van Gool, Luc [1 ,2 ,3 ]
机构：
[1] Katholieke Univ Leuven, Leuven, Belgium
[2] Swiss Fed Inst Technol, Zurich, Switzerland
[3] INSAIT Sofia, Sofia, Bulgaria
来源：
PROGRESS IN PATTERN RECOGNITION, IMAGE ANALYSIS, COMPUTER VISION, AND APPLICATIONS, CIARP 2023, PT I | 2024年 / 14469卷
关键词：
Visual odometry; Online adaptation; Experience replay;
D O I：
10.1007/978-3-031-49018-7_48
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Online adaptation is a promising paradigm that enables dynamic adaptation to new environments. In recent years, there has been a growing interest in exploring online adaptation for various problems, including visual odometry, a crucial task in robotics, autonomous systems, and driver assistance applications. In this work, we leverage experience replay, a potent technique for enhancing online adaptation, to explore the replay-based online adaptation for unsupervised deep visual odometry. Our experiments reveal a remarkable performance boost compared to the non-adapted model. Furthermore, we conduct a comparative analysis against established methods, demonstrating competitive results that showcase the potential of online adaptation in advancing visual odometry.
引用
收藏
页码：674 / 684
页数：11
相关论文
共 50 条
  • [1] Generalizing to the Open World: Deep Visual Odometry with Online Adaptation
    Li, Shunkai
    Wu, Xin
    Cao, Yingdian
    Zha, Hongbin
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 13179 - 13188
  • [2] Self-Supervised Deep Visual Odometry with Online Adaptation
    Li, Shunkai
    Wang, Xin
    Cao, Yingdian
    Xue, Fei
    Yan, Zike
    Zha, Hongbin
    2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, : 6338 - 6347
  • [3] A pproximate and M emorize (A&M) : Settling opposing views in replay-based continuous unsupervised domain adaptation
    Hassan, Mohamed Abubakr
    Elmallah, Ramy
    Lee, Chi-Guhn
    KNOWLEDGE-BASED SYSTEMS, 2024, 293
  • [4] Monocular Visual Odometry Using Unsupervised Deep Learning
    Liu, Fanning
    Liu, Zhenghua
    Wu, Qian
    2019 CHINESE AUTOMATION CONGRESS (CAC2019), 2019, : 3274 - 3279
  • [5] Unsupervised Deep Learning-Based RGB-D Visual Odometry
    Liu, Qiang
    Zhang, Haidong
    Xu, Yiming
    Wang, Li
    APPLIED SCIENCES-BASEL, 2020, 10 (16):
  • [6] Deep Online Correction for Monocular Visual Odometry
    Zhang, Jiaxin
    Sui, Wei
    Wang, Xinggang
    Meng, Wenming
    Zhu, Hongmei
    Zhang, Qian
    2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021), 2021, : 14396 - 14402
  • [7] UnDeepVO: Monocular Visual Odometry through Unsupervised Deep Learning
    Li, Ruihao
    Wang, Sen
    Long, Zhiqiang
    Gu, Dongbing
    2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2018, : 7286 - 7291
  • [8] Using Unsupervised Deep Learning Technique for Monocular Visual Odometry
    Liu, Qiang
    Li, Ruihao
    Hu, Huosheng
    Gu, Dongbing
    IEEE ACCESS, 2019, 7 : 18076 - 18088
  • [9] Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery
    Shamwell, E. Jared
    Lindgren, Kyle
    Leung, Sarah
    Nothwang, William D.
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2020, 42 (10) : 2478 - 2493
  • [10] DEEP UNSUPERVISED LEARNING FOR SIMULTANEOUS VISUAL ODOMETRY AND DEPTH ESTIMATION
    Lu, Yawen
    Lu, Guoyu
    2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2019, : 2571 - 2575
12345