Dense-depth map guided deep Lidar-Visual Odometry with Sparse Point Clouds and Images
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912494566506496 |
|---|---|
| author | Huang, JunYing Xu, Ao Yong, DongSun Li, KeRen Wang, YuanFeng Qin, Qi |
| author_facet | Huang, JunYing Xu, Ao Yong, DongSun Li, KeRen Wang, YuanFeng Qin, Qi |
| contents | Odometry is a critical task for autonomous systems for self-localization and navigation. We propose a novel LiDAR-Visual odometry framework that integrates LiDAR point clouds and images for accurate and robust pose estimation. Our method utilizes a dense-depth map estimated from point clouds and images through depth completion, and incorporates a multi-scale feature extraction network with attention mechanisms, enabling adaptive depth-aware representations. Furthermore, we leverage dense depth information to refine flow estimation and mitigate errors in occlusion-prone regions. Our hierarchical pose refinement module optimizes motion estimation progressively, ensuring robust predictions against dynamic environments and scale ambiguities. Comprehensive experiments on the KITTI odometry benchmark demonstrate that our approach achieves similar or superior accuracy and robustness compared to state-of-the-art visual and LiDAR odometry methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_15496 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dense-depth map guided deep Lidar-Visual Odometry with Sparse Point Clouds and Images Huang, JunYing Xu, Ao Yong, DongSun Li, KeRen Wang, YuanFeng Qin, Qi Computer Vision and Pattern Recognition Machine Learning Robotics Odometry is a critical task for autonomous systems for self-localization and navigation. We propose a novel LiDAR-Visual odometry framework that integrates LiDAR point clouds and images for accurate and robust pose estimation. Our method utilizes a dense-depth map estimated from point clouds and images through depth completion, and incorporates a multi-scale feature extraction network with attention mechanisms, enabling adaptive depth-aware representations. Furthermore, we leverage dense depth information to refine flow estimation and mitigate errors in occlusion-prone regions. Our hierarchical pose refinement module optimizes motion estimation progressively, ensuring robust predictions against dynamic environments and scale ambiguities. Comprehensive experiments on the KITTI odometry benchmark demonstrate that our approach achieves similar or superior accuracy and robustness compared to state-of-the-art visual and LiDAR odometry methods. |
| title | Dense-depth map guided deep Lidar-Visual Odometry with Sparse Point Clouds and Images |
| topic | Computer Vision and Pattern Recognition Machine Learning Robotics |
| url | https://arxiv.org/abs/2507.15496 |