Comparisons of different label assignment methods. H and W are height and width of feature map, respectively, K is number of object categories. Previous works on one-stage object detection assign labels by only position cost, such as (a) box IoU or (b) point distance between sample and ground-truth. In our method, however, (c) classification cost is additionally introduced. We discover that classification cost is the key to the success of end-to-end. Without classification cost, only location cost leads to redundant boxes of high confidence scores in inference, making NMS post-processing a necessary component.
arxiv: OneNet: Towards End-to-End One-Stage Object Detection
paper: What Makes for End-to-End Object Detection?
We provide two models
Method | inf_time | train_time | box AP | download |
---|---|---|---|---|
R18_dcn | 109 FPS | 20h | 29.9 | model | log |
R18_nodcn | 138 FPS | 13h | 27.7 | model | log |
R50_dcn | 67 FPS | 36h | 35.7 | model | log |
R50_nodcn | 73 FPS | 29h | 32.7 | model | log |
R50_RetinaNet | 26 FPS | 31h | 37.5 | model | log |
R50_FCOS | 27 FPS | 21h | 38.9 | model | log |
If download link is invalid, models and logs are also available in Github Release and Baidu Drive by code nhr8.
Method | inf_time | train_time | AP50 | mMR | recall | download |
---|---|---|---|---|---|---|
R50_RetinaNet | 26 FPS | 11.5h | 90.9 | 48.8 | 98.0 | model | log |
R50_FCOS | 27 FPS | 4.5h | 90.6 | 48.6 | 97.7 | model | log |
If download link is invalid, models and logs are also available in Github Release and Baidu Drive by code nhr8.
The codebases are built on top of Detectron2 and DETR.
Install and build libs
git clone https://github.com/PeizeSun/OneNet.git
cd OneNet
python setup.py build develop
Link coco dataset path to OneNet/datasets/coco
mkdir -p datasets/coco
ln -s /path_to_coco_dataset/annotations datasets/coco/annotations
ln -s /path_to_coco_dataset/train2017 datasets/coco/train2017
ln -s /path_to_coco_dataset/val2017 datasets/coco/val2017
Train OneNet
python projects/OneNet/train_net.py --num-gpus 8 \
--config-file projects/OneNet/configs/onenet.res50.dcn.yaml
Evaluate OneNet
python projects/OneNet/train_net.py --num-gpus 8 \
--config-file projects/OneNet/configs/onenet.res50.dcn.yaml \
--eval-only MODEL.WEIGHTS path/to/model.pth
Visualize OneNet
python demo/demo.py\
--config-file projects/OneNet/configs/onenet.res50.dcn.yaml \
--input path/to/images --output path/to/save_images --confidence-threshold 0.4 \
--opts MODEL.WEIGHTS path/to/model.pth
OneNet is released under MIT License.
If you use OneNet in your research or wish to refer to the baseline results published here, please use the following BibTeX entries:
@InProceedings{peize2020onenet,
title = {What Makes for End-to-End Object Detection?},
author = {Sun, Peize and Jiang, Yi and Xie, Enze and Shao, Wenqi and Yuan, Zehuan and Wang, Changhu and Luo, Ping},
booktitle = {Proceedings of the 38th International Conference on Machine Learning},
pages = {9934--9944},
year = {2021},
volume = {139},
series = {Proceedings of Machine Learning Research},
publisher = {PMLR},
}