Filters








3 Hits in 3.4 sec

SVT-Net: Super Light-Weight Sparse Voxel Transformer for Large Scale Place Recognition [article]

Zhaoxin Fan, Zhenbo Song, Hongyan Liu, Zhiwu Lu, Jun He, Xiaoyong Du
2021 arXiv   pre-print
To overcome these challenges, we propose a super light-weight network model termed SVT-Net for large scale place recognition.  ...  Point cloud-based large scale place recognition is fundamental for many applications like Simultaneous Localization and Mapping (SLAM).  ...  Conclusions In this paper, we proposed a super light-weight network for large scale place recognition named SVT-Net.  ... 
arXiv:2105.00149v4 fatcat:3n4er3fiynh3fhuytbsacwgpde

SVT-Net: Super Light-Weight Sparse Voxel Transformer for Large Scale Place Recognition

Zhaoxin Fan, Zhenbo Song, Hongyan Liu, Zhiwu Lu, Jun He, Xiaoyong Du
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Moreover, the model size also becomes a serious shackle for their wide applications. To overcome these challenges, we propose a super light-weight network model termed SVT-Net.  ...  Point cloud-based large scale place recognition is the spine of them.  ...  Conclusions In this paper, we proposed a super light-weight network for large scale place recognition named SVT-Net.  ... 
doi:10.1609/aaai.v36i1.19934 fatcat:2xsum7dydjbkrid6dq2k7xv2ju

Transformers in 3D Point Clouds: A Survey [article]

Dening Lu, Qian Xie, Mingqiang Wei, Kyle Gao, Linlin Xu, Jonathan Li
2022 arXiv   pre-print
For the first time, we provided a comprehensive overview of increasingly popular Transformers for 3D point cloud analysis.  ...  However, how do Transformers cope with the irregularity and unordered nature of point clouds? How suitable are Transformers for different 3D representations (e.g., point- or voxel-based)?  ...  [48] presented Super light-weight Sparse Voxel Transformer (SVT-Net) for large scale place recognition.  ... 
arXiv:2205.07417v2 fatcat:hiyzlryrtbbrppf5pr2lhvkaeq