📖 VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection (CVPR 2026)
🔥Please star VGGT-Det ⭐ and share it. Thanks🔥
Yang Cao*, Feize Wu*, Dave Zhenyu Chen, Yingji Zhong, Lanqing Hong, Dan Xu#
The Hong Kong University of Science and Technology
Huawei Sun Yat-Sen University
🚩 Updates
☑ The training and testing codes on Scannet are released.
☑ The pretrained models and training logs are released at here.
☑ The processed ARKitScenes datasets are released at here.
☑ The processed ScanNet datasets are released at here.
☑ The paper is released at Hugging Face and Arxiv.
☑ Our VGGT-Det is accepted by CVPR 2026. The paper and codes will be released soon.
- Install mmdetection3d
- Install torch-scatter:
pip install torch-scatter==2.1.2 -f https://data.pyg.org/whl/torch-2.1.0%2Bcu118.html
Please download the datasets from here.
Then run for the downloaded *.tar file:
bash data_preparation.sh
Download the pretrained models here. Then run:
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 bash tools/dist_test.sh projects/VGGTDet/config/vggtdet_scannet.py VGGT-Det-Pretrained-Models/ScanNet/epoch_180.pth 8
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 bash tools/dist_train.sh projects/VGGTDet/config/vggtdet_scannet.py 8
If VGGT-Det is helpful, please cite:
@inproceedings{cao2026vggtdet,
title={VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection},
author={Cao, Yang and Wu, Feize and Dave Chen, Zhenyu and Zhong, Yingji and Hong, Lanqing and Xu, Dan},
booktitle={CVPR},
year={2026}
}
If you have any question, please email yangcao.cs@gmail.com.
Appreciate the following works for their great contributions:
VGGT: Inspire our study for Sensor-Geometry-Free 3DDet.
MVSDet, NeRF-Det and MMDet3D: Serve as the foundation for our codes.
ScanNet and ARKitScenes: Serve as the datasets for training and evaluation.