Two-stage object detectors that use class-agnostic one-stage detectors as the proposal network.
Probabilistic two-stage detection,
Xingyi Zhou, Vladlen Koltun, Philipp Krähenbühl,
arXiv technical report (arXiv 2103.07461)
Contact: zhouxy@cs.utexas.edu. Any questions or discussions are welcomed!
-
Two-stage CenterNet: First stage estimates object probabilities, second stage conditionally classifies objects.
-
Resulting detector is faster and more accurate than both traditional two-stage detectors (fewer proposals required), and one-stage detectors (lighter first stage head).
-
Our best model achieves 56.4 mAP on COCO test-dev.
-
This repo also includes a detectron2-based CenterNet implementation with better accuracy (42.5 mAP at 70FPS) and a new FPN version of CenterNet (40.2 mAP with Res50_1x).
All models are trained with multi-scale training, and tested with a single scale. The FPS is tested on a Titan RTX GPU. More models and details can be found in the MODEL_ZOO.
| Model | COCO val mAP | FPS |
|---|---|---|
| CenterNet-S4_DLA_8x | 42.5 | 71 |
| CenterNet2_R50_1x | 42.9 | 24 |
| CenterNet2_X101-DCN_2x | 49.9 | 8 |
| CenterNet2_R2-101-DCN-BiFPN_4x+4x_1560_ST | 56.1 | 5 |
| CenterNet2_DLA-BiFPN-P5_24x_ST | 49.2 | 38 |
| Model | val mAP box |
|---|---|
| CenterNet2_R50_1x | 26.5 |
| CenterNet2_FedLoss_R50_1x | 28.3 |
| Model | val mAP |
|---|---|
| CenterNet2_R50_1x | 22.6 |
Our project is developed on detectron2. Please follow the official detectron2 installation.
We use the default detectron2 demo script. To run inference on an image folder using our pre-trained model, run
python demo.py --config-file configs/CenterNet2_R50_1x.yaml --input path/to/image/ --opts MODEL.WEIGHTS models/CenterNet2_R50_1x.pth
Please check detectron2 GETTING_STARTED.md for running evaluation and training. Our config files are under configs and the pre-trained models are in the MODEL_ZOO.
Our code is under Apache 2.0 license. centernet/modeling/backbone/bifpn_fcos.py are from AdelaiDet, which follows the original non-commercial license.
If you find this project useful for your research, please use the following BibTeX entry.
@inproceedings{zhou2021probablistic,
title={Probabilistic two-stage detection},
author={Zhou, Xingyi and Koltun, Vladlen and Kr{\"a}henb{\"u}hl, Philipp},
booktitle={arXiv preprint arXiv:2103.07461},
year={2021}
}