PeP: a Point enhanced Painting method for unified point cloud tasks
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917599803080704 |
|---|---|
| author | Dong, Zichao Ji, Hang Huang, Xufeng Zhang, Weikun Zhan, Xin Chen, Junbo |
| author_facet | Dong, Zichao Ji, Hang Huang, Xufeng Zhang, Weikun Zhan, Xin Chen, Junbo |
| contents | Point encoder is of vital importance for point cloud recognition. As the very beginning step of whole model pipeline, adding features from diverse sources and providing stronger feature encoding mechanism would provide better input for downstream modules. In our work, we proposed a novel PeP module to tackle above issue. PeP contains two main parts, a refined point painting method and a LM-based point encoder. Experiments results on the nuScenes and KITTI datasets validate the superior performance of our PeP. The advantages leads to strong performance on both semantic segmentation and object detection, in both lidar and multi-modal settings. Notably, our PeP module is model agnostic and plug-and-play. Our code will be publicly available soon. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_07591 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | PeP: a Point enhanced Painting method for unified point cloud tasks Dong, Zichao Ji, Hang Huang, Xufeng Zhang, Weikun Zhan, Xin Chen, Junbo Computer Vision and Pattern Recognition Point encoder is of vital importance for point cloud recognition. As the very beginning step of whole model pipeline, adding features from diverse sources and providing stronger feature encoding mechanism would provide better input for downstream modules. In our work, we proposed a novel PeP module to tackle above issue. PeP contains two main parts, a refined point painting method and a LM-based point encoder. Experiments results on the nuScenes and KITTI datasets validate the superior performance of our PeP. The advantages leads to strong performance on both semantic segmentation and object detection, in both lidar and multi-modal settings. Notably, our PeP module is model agnostic and plug-and-play. Our code will be publicly available soon. |
| title | PeP: a Point enhanced Painting method for unified point cloud tasks |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2310.07591 |