LGU-SLAM: Learnable Gaussian Uncertainty Matching with Deformable Correlation Sampling for Deep Visual SLAM
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909372681027584 |
|---|---|
| author | Huang, Yucheng Ji, Luping Liu, Hudong Ye, Mao |
| author_facet | Huang, Yucheng Ji, Luping Liu, Hudong Ye, Mao |
| contents | Deep visual Simultaneous Localization and Mapping (SLAM) techniques, e.g., DROID, have made significant advancements by leveraging deep visual odometry on dense flow fields. In general, they heavily rely on global visual similarity matching. However, the ambiguous similarity interference in uncertain regions could often lead to excessive noise in correspondences, ultimately misleading SLAM in geometric modeling. To address this issue, we propose a Learnable Gaussian Uncertainty (LGU) matching. It mainly focuses on precise correspondence construction. In our scheme, a learnable 2D Gaussian uncertainty model is designed to associate matching-frame pairs. It could generate input-dependent Gaussian distributions for each correspondence map. Additionally, a multi-scale deformable correlation sampling strategy is devised to adaptively fine-tune the sampling of each direction by a priori look-up ranges, enabling reliable correlation construction. Furthermore, a KAN-bias GRU component is adopted to improve a temporal iterative enhancement for accomplishing sophisticated spatio-temporal modeling with limited parameters. The extensive experiments on real-world and synthetic datasets are conducted to validate the effectiveness and superiority of our method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_23231 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LGU-SLAM: Learnable Gaussian Uncertainty Matching with Deformable Correlation Sampling for Deep Visual SLAM Huang, Yucheng Ji, Luping Liu, Hudong Ye, Mao Computer Vision and Pattern Recognition Deep visual Simultaneous Localization and Mapping (SLAM) techniques, e.g., DROID, have made significant advancements by leveraging deep visual odometry on dense flow fields. In general, they heavily rely on global visual similarity matching. However, the ambiguous similarity interference in uncertain regions could often lead to excessive noise in correspondences, ultimately misleading SLAM in geometric modeling. To address this issue, we propose a Learnable Gaussian Uncertainty (LGU) matching. It mainly focuses on precise correspondence construction. In our scheme, a learnable 2D Gaussian uncertainty model is designed to associate matching-frame pairs. It could generate input-dependent Gaussian distributions for each correspondence map. Additionally, a multi-scale deformable correlation sampling strategy is devised to adaptively fine-tune the sampling of each direction by a priori look-up ranges, enabling reliable correlation construction. Furthermore, a KAN-bias GRU component is adopted to improve a temporal iterative enhancement for accomplishing sophisticated spatio-temporal modeling with limited parameters. The extensive experiments on real-world and synthetic datasets are conducted to validate the effectiveness and superiority of our method. |
| title | LGU-SLAM: Learnable Gaussian Uncertainty Matching with Deformable Correlation Sampling for Deep Visual SLAM |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.23231 |