Label-Efficient 3D Object Detection For Road-Side Units
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910403532947456 |
|---|---|
| author | Dao, Minh-Quan Caesar, Holger Berrio, Julie Stephany Shan, Mao Worrall, Stewart Frémont, Vincent Malis, Ezio |
| author_facet | Dao, Minh-Quan Caesar, Holger Berrio, Julie Stephany Shan, Mao Worrall, Stewart Frémont, Vincent Malis, Ezio |
| contents | Occlusion presents a significant challenge for safety-critical applications such as autonomous driving. Collaborative perception has recently attracted a large research interest thanks to the ability to enhance the perception of autonomous vehicles via deep information fusion with intelligent roadside units (RSU), thus minimizing the impact of occlusion. While significant advancement has been made, the data-hungry nature of these methods creates a major hurdle for their real-world deployment, particularly due to the need for annotated RSU data. Manually annotating the vast amount of RSU data required for training is prohibitively expensive, given the sheer number of intersections and the effort involved in annotating point clouds. We address this challenge by devising a label-efficient object detection method for RSU based on unsupervised object discovery. Our paper introduces two new modules: one for object discovery based on a spatial-temporal aggregation of point clouds, and another for refinement. Furthermore, we demonstrate that fine-tuning on a small portion of annotated data allows our object discovery models to narrow the performance gap with, or even surpass, fully supervised models. Extensive experiments are carried out in simulated and real-world datasets to evaluate our method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_06256 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Label-Efficient 3D Object Detection For Road-Side Units Dao, Minh-Quan Caesar, Holger Berrio, Julie Stephany Shan, Mao Worrall, Stewart Frémont, Vincent Malis, Ezio Computer Vision and Pattern Recognition Robotics Occlusion presents a significant challenge for safety-critical applications such as autonomous driving. Collaborative perception has recently attracted a large research interest thanks to the ability to enhance the perception of autonomous vehicles via deep information fusion with intelligent roadside units (RSU), thus minimizing the impact of occlusion. While significant advancement has been made, the data-hungry nature of these methods creates a major hurdle for their real-world deployment, particularly due to the need for annotated RSU data. Manually annotating the vast amount of RSU data required for training is prohibitively expensive, given the sheer number of intersections and the effort involved in annotating point clouds. We address this challenge by devising a label-efficient object detection method for RSU based on unsupervised object discovery. Our paper introduces two new modules: one for object discovery based on a spatial-temporal aggregation of point clouds, and another for refinement. Furthermore, we demonstrate that fine-tuning on a small portion of annotated data allows our object discovery models to narrow the performance gap with, or even surpass, fully supervised models. Extensive experiments are carried out in simulated and real-world datasets to evaluate our method. |
| title | Label-Efficient 3D Object Detection For Road-Side Units |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2404.06256 |