YOLOF

You Only Look One-level Feature (YOLOF), CVPR2021.

Getting Started

  • Install cvpods by following the installation guidance.
  • Install mish-cuda to speed up the training and inference when using CSPDarkNet-53 as the backbone (optional)
    git clone https://github.com/thomasbrandon/mish-cuda
cd mish-cuda
python setup.py build install
cd ..
  • Download the pretrained model in OneDrive or in the Baidu Cloud with code qr6o to train with the CSPDarkNet-53 backbone (optional)
    mkdir pretrained_models
# download the `cspdarknet53.pth` to the `pretrained_models` directory
  • Train
    cd playground/detection/coco/yolof/yolof.res50.C5.1x
pods_train --num-gpus 8
  • Test
    cd playground/detection/coco/yolof/yolof.res50.C5.1x
pods_test --num-gpus 8 MODEL.WEIGHTS /path/to/checkpoint_file

Main results

The models listed below can be found in this onedrive link or in the BaiduCloud link with code qr6o. The FPS is tested on a 2080Ti GPU.
More models will be available in the near future.

Model COCO val mAP FPS
YOLOF_R_50_C5_1x 37.7 32
YOLOF_R_50_DC5_1x 39.2 20
YOLOF_R_101_C5_1x 39.8 21
YOLOF_R_101_DC5_1x 40.5 15
YOLOF_CSP_D_53_DC5_3x 41.2 39
YOLOF_CSP_D_53_DC5_9x 42.8 39

Citation

If you find this project useful for your research, please use the following BibTeX entry.

@inproceedings{chen2021you,
  title={You Only Look One-level Feature},
  author={Chen, Qiang and Wang, Yingming and Yang, Tong and Zhang, Xiangyu and Cheng, Jian and Sun, Jian},
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
  year={2021}
}
View Github