Skip to content

[CVPR 2023] FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detection

Notifications You must be signed in to change notification settings

Robertwyq/Frustum

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FrustumFormer

[CVPR 2023] FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detection.

Authors: Yuqi Wang, Yuntao Chen, Zhaoxiang Zhang

Video, Paper

alt text

News

  • [2023/3/26] We release the camera ready version.
  • [2023/2/28] FrustumFormer was accepted by CVPR 2023.

Catalog

  • Initialization

Getting Started

Model Zoo

NuScenes Val

Backbone Method Epoch Img mAP NDS memroy Config Download
R101-DCN FrustumFormer 24 640*1600 0.472 0.553 40G Frustum_r101

NuScenes Test

Backbone Method Epoch Img mAP NDS memroy Config Download
VoV99 FrustumFormer 24 640*1600 0.530 0.595 40G Frustum_vov99

Bibtex

If this work is helpful for your research, please consider citing the following BibTeX entry.

@inproceedings{wang2023frustumformer,
  title={FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detection},
  author={Wang, Yuqi and Chen, Yuntao and Zhang, Zhaoxiang},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={5096--5105},
  year={2023}
}

Acknowledgement

Many thanks to the following open-source projects:

About

[CVPR 2023] FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detection

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages