Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915517652008960 |
|---|---|
| author | Kanai, Takayuki Vasiljevic, Igor Guizilini, Vitor Shintani, Kazuhiro |
| author_facet | Kanai, Takayuki Vasiljevic, Igor Guizilini, Vitor Shintani, Kazuhiro |
| contents | Monocular visual odometry is a key technology in various autonomous systems. Traditional feature-based methods suffer from failures due to poor lighting, insufficient texture, and large motions. In contrast, recent learning-based dense SLAM methods exploit iterative dense bundle adjustment to address such failure cases, and achieve robust and accurate localization in a wide variety of real environments, without depending on domain-specific supervision. However, despite its potential, the methods still struggle with scenarios involving large motion and object dynamics. In this study, we diagnose key weaknesses in a popular learning-based dense SLAM model (DROID-SLAM) by analyzing major failure cases on outdoor benchmarks and exposing various shortcomings of its optimization process. We then propose the use of self-supervised priors leveraging a frozen large-scale pre-trained monocular depth estimator to initialize the dense bundle adjustment process, leading to robust visual odometry without the need to fine-tune the SLAM backbone. Despite its simplicity, the proposed method demonstrates significant improvements on KITTI odometry, as well as the challenging DDAD benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_00929 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry Kanai, Takayuki Vasiljevic, Igor Guizilini, Vitor Shintani, Kazuhiro Computer Vision and Pattern Recognition Robotics Monocular visual odometry is a key technology in various autonomous systems. Traditional feature-based methods suffer from failures due to poor lighting, insufficient texture, and large motions. In contrast, recent learning-based dense SLAM methods exploit iterative dense bundle adjustment to address such failure cases, and achieve robust and accurate localization in a wide variety of real environments, without depending on domain-specific supervision. However, despite its potential, the methods still struggle with scenarios involving large motion and object dynamics. In this study, we diagnose key weaknesses in a popular learning-based dense SLAM model (DROID-SLAM) by analyzing major failure cases on outdoor benchmarks and exposing various shortcomings of its optimization process. We then propose the use of self-supervised priors leveraging a frozen large-scale pre-trained monocular depth estimator to initialize the dense bundle adjustment process, leading to robust visual odometry without the need to fine-tune the SLAM backbone. Despite its simplicity, the proposed method demonstrates significant improvements on KITTI odometry, as well as the challenging DDAD benchmark. |
| title | Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2406.00929 |