Views
No views yet
1git clone https://github.com/wlfeng0509/QuantVGGT.git
2cd QuantVGGT
3pip install -r requirements.txt
4pip install -r requirements_demo.txt1python Quant_VGGT/vggt/evaluation/make_calibation.py \
2 --model_path VGGT-1B/model_tracker_fixed_e20.pt \
3 --co3d_dir co3d_datasets/ \
4 --co3d_anno_dir co3d_v2_annotations/ \
5 --seed 0 \
6 --cache_path all_calib_data.pt \
7 --save_path calib_data.pt \
8 --class_mode all \
9 --kmeans_n 6 \
10 --kmeans_m 71python Quant_VGGT/vggt/evaluation/run_co3d.py \
2 --model_path Quant_VGGT/VGGT-1B/model_tracker_fixed_e20.pt \
3 --co3d_dir co3d_datasets/ \
4 --co3d_anno_dir co3d_v2_annotations/ \
5 --dtype quarot_w4a4 \
6 --seed 0 \
7 --lac \
8 --lwc \
9 --cache_path calib_data.pt \
10 --class_mode all \
11 --exp_name a44_uqant \
12 --resume_qs1@article{feng2025quantized,
2 title={Quantized Visual Geometry Grounded Transformer},
3 author={Feng, Weilun and Qin, Haotong and Wu, Mingqiang and Yang, Chuanguang and Li, Yuqi and Li, Xiangqi and An, Zhulin and Huang, Libo and Zhang, Yulun and Magno, Michele and others},
4 journal={arXiv preprint arXiv:2509.21302},
5 year={2025}
6}