OctreeOcc: Efficient and Multi-Granularity Occupancy Prediction Using Octree Queries
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909141068414976 |
|---|---|
| author | Lu, Yuhang Zhu, Xinge Wang, Tai Ma, Yuexin |
| author_facet | Lu, Yuhang Zhu, Xinge Wang, Tai Ma, Yuexin |
| contents | Occupancy prediction has increasingly garnered attention in recent years for its fine-grained understanding of 3D scenes. Traditional approaches typically rely on dense, regular grid representations, which often leads to excessive computational demands and a loss of spatial details for small objects. This paper introduces OctreeOcc, an innovative 3D occupancy prediction framework that leverages the octree representation to adaptively capture valuable information in 3D, offering variable granularity to accommodate object shapes and semantic regions of varying sizes and complexities. In particular, we incorporate image semantic information to improve the accuracy of initial octree structures and design an effective rectification mechanism to refine the octree structure iteratively. Our extensive evaluations show that OctreeOcc not only surpasses state-of-the-art methods in occupancy prediction, but also achieves a 15%-24% reduction in computational overhead compared to dense-grid-based methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_03774 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | OctreeOcc: Efficient and Multi-Granularity Occupancy Prediction Using Octree Queries Lu, Yuhang Zhu, Xinge Wang, Tai Ma, Yuexin Computer Vision and Pattern Recognition Occupancy prediction has increasingly garnered attention in recent years for its fine-grained understanding of 3D scenes. Traditional approaches typically rely on dense, regular grid representations, which often leads to excessive computational demands and a loss of spatial details for small objects. This paper introduces OctreeOcc, an innovative 3D occupancy prediction framework that leverages the octree representation to adaptively capture valuable information in 3D, offering variable granularity to accommodate object shapes and semantic regions of varying sizes and complexities. In particular, we incorporate image semantic information to improve the accuracy of initial octree structures and design an effective rectification mechanism to refine the octree structure iteratively. Our extensive evaluations show that OctreeOcc not only surpasses state-of-the-art methods in occupancy prediction, but also achieves a 15%-24% reduction in computational overhead compared to dense-grid-based methods. |
| title | OctreeOcc: Efficient and Multi-Granularity Occupancy Prediction Using Octree Queries |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2312.03774 |