Improving Generalization Ability for 3D Object Detection by Learning Sparsity-invariant Features
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916597386444800 |
|---|---|
| author | Lu, Hsin-Cheng Lin, Chung-Yi Hsu, Winston H. |
| author_facet | Lu, Hsin-Cheng Lin, Chung-Yi Hsu, Winston H. |
| contents | In autonomous driving, 3D object detection is essential for accurately identifying and tracking objects. Despite the continuous development of various technologies for this task, a significant drawback is observed in most of them-they experience substantial performance degradation when detecting objects in unseen domains. In this paper, we propose a method to improve the generalization ability for 3D object detection on a single domain. We primarily focus on generalizing from a single source domain to target domains with distinct sensor configurations and scene distributions. To learn sparsity-invariant features from a single source domain, we selectively subsample the source data to a specific beam, using confidence scores determined by the current detector to identify the density that holds utmost importance for the detector. Subsequently, we employ the teacher-student framework to align the Bird's Eye View (BEV) features for different point clouds densities. We also utilize feature content alignment (FCA) and graph-based embedding relationship alignment (GERA) to instruct the detector to be domain-agnostic. Extensive experiments demonstrate that our method exhibits superior generalization capabilities compared to other baselines. Furthermore, our approach even outperforms certain domain adaptation methods that can access to the target domain data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_02322 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Improving Generalization Ability for 3D Object Detection by Learning Sparsity-invariant Features Lu, Hsin-Cheng Lin, Chung-Yi Hsu, Winston H. Computer Vision and Pattern Recognition Robotics In autonomous driving, 3D object detection is essential for accurately identifying and tracking objects. Despite the continuous development of various technologies for this task, a significant drawback is observed in most of them-they experience substantial performance degradation when detecting objects in unseen domains. In this paper, we propose a method to improve the generalization ability for 3D object detection on a single domain. We primarily focus on generalizing from a single source domain to target domains with distinct sensor configurations and scene distributions. To learn sparsity-invariant features from a single source domain, we selectively subsample the source data to a specific beam, using confidence scores determined by the current detector to identify the density that holds utmost importance for the detector. Subsequently, we employ the teacher-student framework to align the Bird's Eye View (BEV) features for different point clouds densities. We also utilize feature content alignment (FCA) and graph-based embedding relationship alignment (GERA) to instruct the detector to be domain-agnostic. Extensive experiments demonstrate that our method exhibits superior generalization capabilities compared to other baselines. Furthermore, our approach even outperforms certain domain adaptation methods that can access to the target domain data. |
| title | Improving Generalization Ability for 3D Object Detection by Learning Sparsity-invariant Features |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2502.02322 |