YOLOU：United, Study and easier to Deploy

​ The purpose of our creation of YOLOU is to better learn the algorithms of the YOLO series and pay tribute to our predecessors.

​ Here "U" means United, mainly to gather more algorithms about the YOLO series through this project, so that friends can better learn the knowledge of object detection. At the same time, in order to better apply AI technology, YOLOU will also join The corresponding Deploy technology will accelerate the implementation of the algorithms we have learned and realize the value.

At present, the YOLO series algorithms mainly included in YOLOU are:

Anchor-base: YOLOv3, YOLOv4, YOLOv5, YOLOv5-Lite, YOLOv7, YOLOv5-TPH, YOLO-Fastest v2, YOLO-LF, YOLO-SA, YOLOR, YOLOv5-SPD

Anchor-Free: YOLOv6-v1, YOLOv6-v2, YOLOX, YOLOE, YOLOX-Lite, FastestDet

Face-Detection: YOLOv5-Face, YOLOFace-v2

Segmentation: YOLOv5-Segment

KeyPoint: YOLOv7-Keypoint

Classfication: ResNet, DarkNet,......

Comparison of ablation experiment results

Model	size(pixels)	mAP@.5	mAP@.5:95	Parameters(M)	GFLOPs	TensorRT-FP32(b16) ms/fps	TensorRT-FP16(b16) ms/fps
YOLOv5n	640	45.7	28.0	1.9	4.5	0.95/1054.64	0.61/1631.64
YOLOv5s	640	56.8	37.4	7.2	16.5	1.7/586.8	0.84/1186.42
YOLOv5m	640	64.1	45.4	21.2	49.0	4.03/248.12	1.42/704.20
YOLOv5l	640	67.3	49.0	46.5	109.1
YOLOv5x	640	68.9	50.7	86.7	205.7
YOLOv6-T	640
YOLOv6-n	640
YOLOv6	640	58.4	39.8	20.4	28.8	3.06/326.93	1.27/789.51
YOLOv7	640	69.7	51.4	37.6	53.1	8.18/113.88	1.97/507.55
YOLOv7-X	640	71.2	53.7	71.3	95.1
YOLOv7-W6	1280	72.6	54.9
YOLOv7-E6	1280	73.5	56.0
YOLOv7-D6	1280	74.0	56.6
YOLOv7-E6E	1280	74.4	56.8
YOLOX-s	640	59.0	39.2	8.1	10.8	2.11/473.78	0.89/1127.67
YOLOX-m	640	63.8	44.5	23.3	31.2	4.94/202.43	1.58/632.48
YOLOX-l	640			54.1	77.7
YOLOX-x	640			104.5	156.2
v5-Lite-e	320	35.1		0.78	0.73	0.55/1816.10	0.49/2048.47
v5-Lite-s	416	42.0	25.2	1.64	1.66	0.72/1384.76	0.64/1567.36
v5-Lite-c	512	50.9	32.5	4.57	5.92	1.18/850.03	0.80/1244.20
v5-Lite-g	640	57.6	39.1	5.39	15.6	1.85/540.90	1.09/916.69
X-Lite-e	320	36.4	21.2	2.53	1.58	0.65/1547.58	0.46/2156.38
X-Lite-s	416	Training…	Training…	3.36	2.90
X-Lite-c	512	Training…	Training…	6.25	5.92
X-Lite-g	640	58.3	40.7	7.30	12.91	2.15/465.19	1.01/990.69

How to use

Install

git clone https://github.com/jizhishutong/YOLOU
cd YOLOU
pip install -r requirements.txt

Training

python train_det.py --mode yolov6 --data coco.yaml --cfg yolov6.yaml --weights yolov6.pt --batch-size 32

Detect

python detect_det.py --source 0  # webcam
                            file.jpg  # image 
                            file.mp4  # video
                            path/  # directory
                            path/*.jpg  # glob
                            'https://youtu.be/NUsoVlDFqZg'  # YouTube
                            'rtsp://example.com/media.mp4'  # RTSP, RTMP, HTTP stream

Re-parameterization

See reparameterization.ipynb

Pose estimation

yolov7-w6-pose.pt

See keypoint.ipynb.

Detect Inference Result

Segmentation Inference Result

KeyPoint Inference Result

Face-Detect Inference Result

DataSet

train: ../coco/images/train2017/
val: ../coco/images/val2017/

├── images            # xx.jpg example
│   ├── train2017        
│   │   ├── 000001.jpg
│   │   ├── 000002.jpg
│   │   └── 000003.jpg
│   └── val2017         
│       ├── 100001.jpg
│       ├── 100002.jpg
│       └── 100003.jpg
└── labels             # xx.txt example      
    ├── train2017       
    │   ├── 000001.txt
    │   ├── 000002.txt
    │   └── 000003.txt
    └── val2017         
        ├── 100001.txt
        ├── 100002.txt
        └── 100003.txt

Export ONNX

Expand

python export.py --weights ./weights/yolov6/yolov6s.pt --include onnx

​ In order to facilitate the deployment and implementation of friends here, all models included in YOLOU have been processed to a certain extent, and their pre- and post-processing codes can be used in one set, because the format and output results of the ONNX files they export are consistent.

YOLOv5

YOLOv6

YOLOv7

YOLOX

YOLOv5-Lite

YOLOX-Lite

YOLO-Fastest v2

Acknowledgements

Expand

https://github.com/ultralytics/yolov5

https://github.com/WongKinYiu/yolor

https://github.com/ppogg/YOLOv5-Lite

https://github.com/WongKinYiu/yolov7

https://github.com/meituan/YOLOv6

https://github.com/ultralytics/yolov3

https://github.com/Megvii-BaseDetection/YOLOX

https://github.com/WongKinYiu/ScaledYOLOv4

https://github.com/WongKinYiu/PyTorch_YOLOv4

https://github.com/WongKinYiu/yolor

https://github.com/shouxieai/tensorRT_Pro

https://github.com/Tencent/ncnn

https://github.com/Gumpest/YOLOv5-Multibackbone-Compression

https://github.com/positive666/yolov5_research

https://github.com/cmdbug/YOLOv5_NCNN

https://github.com/OAID/Tengine

https://github.com/meituan/YOLOv6/releases/tag/0.2.0

https://github.com/PaddlePaddle/PaddleDetection

Citing YOLOU

Expand

If you use YOLOU in your research, please cite our work and give a star ⭐:

 @misc{yolou2022,
  title = { YOLOU：United, Study and easier to Deploy},
  author = {ChaucerG},
  year={2022}
}

For more information, Please follow the QR code below：