Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.23910 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909373744283648 |
|---|---|
| author | Durasov, Nikita Mahmood, Rafid Choi, Jiwoong Law, Marc T. Lucas, James Fua, Pascal Alvarez, Jose M. |
| author_facet | Durasov, Nikita Mahmood, Rafid Choi, Jiwoong Law, Marc T. Lucas, James Fua, Pascal Alvarez, Jose M. |
| contents | 3D object detection is an essential task for computer vision applications in autonomous vehicles and robotics. However, models often struggle to quantify detection reliability, leading to poor performance on unfamiliar scenes. We introduce a framework for quantifying uncertainty in 3D object detection by leveraging an evidential learning loss on Bird's Eye View representations in the 3D detector. These uncertainty estimates require minimal computational overhead and are generalizable across different architectures. We demonstrate both the efficacy and importance of these uncertainty estimates on identifying out-of-distribution scenes, poorly localized objects, and missing (false negative) detections; our framework consistently improves over baselines by 10-20% on average. Finally, we integrate this suite of tasks into a system where a 3D object detector auto-labels driving scenes and our uncertainty estimates verify label correctness before the labels are used to train a second model. Here, our uncertainty-driven verification results in a 1% improvement in mAP and a 1-2% improvement in NDS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_23910 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Uncertainty Estimation for 3D Object Detection via Evidential Learning Durasov, Nikita Mahmood, Rafid Choi, Jiwoong Law, Marc T. Lucas, James Fua, Pascal Alvarez, Jose M. Computer Vision and Pattern Recognition 3D object detection is an essential task for computer vision applications in autonomous vehicles and robotics. However, models often struggle to quantify detection reliability, leading to poor performance on unfamiliar scenes. We introduce a framework for quantifying uncertainty in 3D object detection by leveraging an evidential learning loss on Bird's Eye View representations in the 3D detector. These uncertainty estimates require minimal computational overhead and are generalizable across different architectures. We demonstrate both the efficacy and importance of these uncertainty estimates on identifying out-of-distribution scenes, poorly localized objects, and missing (false negative) detections; our framework consistently improves over baselines by 10-20% on average. Finally, we integrate this suite of tasks into a system where a 3D object detector auto-labels driving scenes and our uncertainty estimates verify label correctness before the labels are used to train a second model. Here, our uncertainty-driven verification results in a 1% improvement in mAP and a 1-2% improvement in NDS. |
| title | Uncertainty Estimation for 3D Object Detection via Evidential Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.23910 |