Boosting 3D Object Detection with Semantic-Aware Multi-Branch Framework
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_ | 1866909644429983744 |
|---|---|
| author | Jing, Hao Wang, Anhong Zhao, Lijun Yang, Yakun Bu, Donghan Zhang, Jing Zhang, Yifan Hou, Junhui |
| author_facet | Jing, Hao Wang, Anhong Zhao, Lijun Yang, Yakun Bu, Donghan Zhang, Jing Zhang, Yifan Hou, Junhui |
| contents | In autonomous driving, LiDAR sensors are vital for acquiring 3D point clouds, providing reliable geometric information. However, traditional sampling methods of preprocessing often ignore semantic features, leading to detail loss and ground point interference in 3D object detection. To address this, we propose a multi-branch two-stage 3D object detection framework using a Semantic-aware Multi-branch Sampling (SMS) module and multi-view consistency constraints. The SMS module includes random sampling, Density Equalization Sampling (DES) for enhancing distant objects, and Ground Abandonment Sampling (GAS) to focus on non-ground points. The sampled multi-view points are processed through a Consistent KeyPoint Selection (CKPS) module to generate consistent keypoint masks for efficient proposal sampling. The first-stage detector uses multi-branch parallel learning with multi-view consistency loss for feature aggregation, while the second-stage detector fuses multi-view data through a Multi-View Fusion Pooling (MVFP) module to precisely predict 3D objects. The experimental results on the KITTI dataset and Waymo Open Dataset show that our method achieves excellent detection performance improvement for a variety of backbones, especially for low-performance backbones with the simple network structures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_05769 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Boosting 3D Object Detection with Semantic-Aware Multi-Branch Framework Jing, Hao Wang, Anhong Zhao, Lijun Yang, Yakun Bu, Donghan Zhang, Jing Zhang, Yifan Hou, Junhui Computer Vision and Pattern Recognition In autonomous driving, LiDAR sensors are vital for acquiring 3D point clouds, providing reliable geometric information. However, traditional sampling methods of preprocessing often ignore semantic features, leading to detail loss and ground point interference in 3D object detection. To address this, we propose a multi-branch two-stage 3D object detection framework using a Semantic-aware Multi-branch Sampling (SMS) module and multi-view consistency constraints. The SMS module includes random sampling, Density Equalization Sampling (DES) for enhancing distant objects, and Ground Abandonment Sampling (GAS) to focus on non-ground points. The sampled multi-view points are processed through a Consistent KeyPoint Selection (CKPS) module to generate consistent keypoint masks for efficient proposal sampling. The first-stage detector uses multi-branch parallel learning with multi-view consistency loss for feature aggregation, while the second-stage detector fuses multi-view data through a Multi-View Fusion Pooling (MVFP) module to precisely predict 3D objects. The experimental results on the KITTI dataset and Waymo Open Dataset show that our method achieves excellent detection performance improvement for a variety of backbones, especially for low-performance backbones with the simple network structures. |
| title | Boosting 3D Object Detection with Semantic-Aware Multi-Branch Framework |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.05769 |