Introduction

YOLOX is an anchor-free version of YOLO, with a simpler design but better performance! It aims to bridge the gap between research and industrial communities. For more details, please refer to our report on Arxiv.

Updates!!

• 【2021/07/20】 We have released our technical report on Arxiv.

Comming soon

• YOLOX-P6 and larger model.
• Objects365 pretrain.
• Transformer modules.
• More features in need.

Benchmark

Standard Models.

Model	size	mAPtest 0.5:0.95	Speed V100 (ms)	Params (M)	FLOPs (G)	weights
YOLOX-s	640	39.6	9.8	9.0	26.8	Download
YOLOX-m	640	46.4	12.3	25.3	73.8	Download
YOLOX-l	640	50.0	14.5	54.2	155.6	Download
YOLOX-x	640	51.2	17.3	99.1	281.9	Download
YOLOX-Darknet53	640	47.4	11.1	63.7	185.3	Download

Light Models.

Model	size	mAPval 0.5:0.95	Params (M)	FLOPs (G)	weights
YOLOX-Nano	416	25.3	0.91	1.08	Download
YOLOX-Tiny	416	31.7	5.06	6.45	Download

Quick Start

Installation

Step1. Install YOLOX.

git clone [email protected]:Megvii-BaseDetection/YOLOX.git
cd YOLOX
pip3 install -U pip && pip3 install -r requirements.txt
pip3 install -v -e .  # or  python3 setup.py develop

Step2. Install apex.

git clone https://github.com/NVIDIA/apex
cd apex
pip3 install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./

Step3. Install pycocotools.

pip3 install cython; pip3 install 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'

Demo

Step1. Download a pretrained model from the benchmark table.

Step2. Use either -n or -f to specify your detector's config. For example:

python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.3 --nms 0.65 --tsize 640 --save_result

or

python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.3 --nms 0.65 --tsize 640 --save_result

Demo for video:

python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth.tar --path /path/to/your/video --conf 0.3 --nms 0.65 --tsize 640 --save_result

Reproduce our results on COCO

Step1. Prepare COCO dataset

cd <YOLOX_HOME>
ln -s /path/to/your/COCO ./datasets/COCO

Step2. Reproduce our results on COCO by specifying -n:

python tools/train.py -n yolox-s -d 8 -b 64 --fp16 -o
                         yolox-m
                         yolox-l
                         yolox-x

• -d: number of gpu devices
• -b: total batch size, the recommended number for -b is num-gpu * 8
• --fp16: mixed precision training

When using -f, the above commands are equivalent to:

python tools/train.py -f exps/default/yolox-s.py -d 8 -b 64 --fp16 -o
                         exps/default/yolox-m.py
                         exps/default/yolox-l.py
                         exps/default/yolox-x.py

Evaluation

We support batch testing for fast evaluation:

python tools/eval.py -n  yolox-s -c yolox_s.pth.tar -b 64 -d 8 --conf 0.001 [--fp16] [--fuse]
                         yolox-m
                         yolox-l
                         yolox-x

• --fuse: fuse conv and bn
• -d: number of GPUs used for evaluation. DEFAULT: All GPUs available will be used.
• -b: total batch size across on all GPUs

To reproduce speed test, we use the following command:

python tools/eval.py -n  yolox-s -c yolox_s.pth.tar -b 1 -d 1 --conf 0.001 --fp16 --fuse
                         yolox-m
                         yolox-l
                         yolox-x

Tutorials

• Training on custom data.

Deployment

1. ONNX export and an ONNXRuntime
2. TensorRT in C++ and Python
3. ncnn in C++ and Java
4. OpenVINO in C++ and Python

Cite YOLOX

If you use YOLOX in your research, please cite our work by using the following BibTeX entry:

 @article{yolox2021,
  title={YOLOX: Exceeding YOLO Series in 2021},
  author={Ge, Zheng and Liu, Songtao and Wang, Feng and Li, Zeming and Sun, Jian},
  journal={arXiv preprint arXiv:2107.08430},
  year={2021}
}