Official YOLOv7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

<a href="https://colab.research.google.com/gist/AlexeyAB/b769f5795e65fdab80086f6cb7940dae/yolov7detection.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>

<div align="center"> <a href="./"> <img src="./figure/performance.png" width="79%"/> </a> </div>

Web Demo

• Integrated into Huggingface Spaces 🤗 using Gradio. Try out the Web Demo

Performance

MS COCO

Model	Test Size	AP<sup>test</sup>	AP<sub>50</sub><sup>test</sup>	AP<sub>75</sub><sup>test</sup>	batch 1 fps	batch 32 average time
YOLOv7	640	51.4%	69.7%	55.9%	161 fps	2.8 ms
YOLOv7-X	640	53.1%	71.2%	57.8%	114 fps	4.3 ms
YOLOv7-W6	1280	54.9%	72.6%	60.1%	84 fps	7.6 ms
YOLOv7-E6	1280	56.0%	73.5%	61.2%	56 fps	12.3 ms
YOLOv7-D6	1280	56.6%	74.0%	61.8%	44 fps	15.0 ms
YOLOv7-E6E	1280	56.8%	74.4%	62.1%	36 fps	18.7 ms

Installation

Docker environment (recommended)

<details><summary> <b>Expand</b> </summary>

shellCopy
# create the docker container, you can change the share memory size if you have more.
nvidia-docker run --name yolov7 -it -v your_coco_path/:/coco/ -v your_code_path/:/yolov7 --shm-size=64g nvcr.io/nvidia/pytorch:21.08-py3

# apt install required packages
apt update
apt install -y zip htop screen libgl1-mesa-glx

# pip install required packages
pip install seaborn thop

# go to code folder
cd /yolov7

</details>

Testing

yolov7.pt yolov7x.pt yolov7-w6.pt yolov7-e6.pt yolov7-d6.pt yolov7-e6e.pt

shellCopy
python test.py --data data/coco.yaml --img 640 --batch 32 --conf 0.001 --iou 0.65 --device 0 --weights yolov7.pt --name yolov7_640_val

You will get the results:

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.51206
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.69730
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.55521
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.35247
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.55937
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.66693
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.38453
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.63765
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.68772
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.53766
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.73549
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.83868

To measure accuracy, download COCO-annotations for Pycocotools to the ./coco/annotations/instances_val2017.json

Training

Data preparation

shellCopy
bash scripts/get_coco.sh

• Download MS COCO dataset images (train, val, test) and labels. If you have previously used a different version of YOLO, we strongly recommend that you delete train2017.cache and val2017.cache files, and redownload labels

Single GPU training

shellCopy
# train p5 models
python train.py --workers 8 --device 0 --batch-size 32 --data data/coco.yaml --img 640 640 --cfg cfg/training/yolov7.yaml --weights '' --name yolov7 --hyp data/hyp.scratch.p5.yaml

# train p6 models
python train_aux.py --workers 8 --device 0 --batch-size 16 --data data/coco.yaml --img 1280 1280 --cfg cfg/training/yolov7-w6.yaml --weights '' --name yolov7-w6 --hyp data/hyp.scratch.p6.yaml

Multiple GPU training

shellCopy
# train p5 models
python -m torch.distributed.launch --nproc_per_node 4 --master_port 9527 train.py --workers 8 --device 0,1,2,3 --sync-bn --batch-size 128 --data data/coco.yaml --img 640 640 --cfg cfg/training/yolov7.yaml --weights '' --name yolov7 --hyp data/hyp.scratch.p5.yaml

# train p6 models
python -m torch.distributed.launch --nproc_per_node 8 --master_port 9527 train_aux.py --workers 8 --device 0,1,2,3,4,5,6,7 --sync-bn --batch-size 128 --data data/coco.yaml --img 1280 1280 --cfg cfg/training/yolov7-w6.yaml --weights '' --name yolov7-w6 --hyp data/hyp.scratch.p6.yaml

Transfer learning

yolov7_training.pt yolov7x_training.pt yolov7-w6_training.pt yolov7-e6_training.pt yolov7-d6_training.pt yolov7-e6e_training.pt

Single GPU finetuning for custom dataset

shellCopy
# finetune p5 models
python train.py --workers 8 --device 0 --batch-size 32 --data data/custom.yaml --img 640 640 --cfg cfg/training/yolov7-custom.yaml --weights 'yolov7_training.pt' --name yolov7-custom --hyp data/hyp.scratch.custom.yaml

# finetune p6 models
python train_aux.py --workers 8 --device 0 --batch-size 16 --data data/custom.yaml --img 1280 1280 --cfg cfg/training/yolov7-w6-custom.yaml --weights 'yolov7-w6_training.pt' --name yolov7-w6-custom --hyp data/hyp.scratch.custom.yaml

Re-parameterization

See reparameterization.ipynb

Inference

On video:

shellCopy
python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --source yourvideo.mp4

On image:

shellCopy
python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --source inference/images/horses.jpg

<div align="center"> <a href="./"> <img src="./figure/horses_prediction.jpg" width="59%"/> </a> </div>

Export

Pytorch to CoreML (and inference on MacOS/iOS) <a href="https://colab.research.google.com/github/WongKinYiu/yolov7/blob/main/tools/YOLOv7CoreML.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>

Pytorch to ONNX with NMS (and inference) <a href="https://colab.research.google.com/github/WongKinYiu/yolov7/blob/main/tools/YOLOv7onnx.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>

shellCopy
python export.py --weights yolov7-tiny.pt --grid --end2end --simplify \
        --topk-all 100 --iou-thres 0.65 --conf-thres 0.35 --img-size 640 640 --max-wh 640

Pytorch to TensorRT with NMS (and inference) <a href="https://colab.research.google.com/github/WongKinYiu/yolov7/blob/main/tools/YOLOv7trt.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>

shellCopy
wget https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-tiny.pt
python export.py --weights ./yolov7-tiny.pt --grid --end2end --simplify --topk-all 100 --iou-thres 0.65 --conf-thres 0.35 --img-size 640 640
git clone https://github.com/Linaom1214/tensorrt-python.git
python ./tensorrt-python/export.py -o yolov7-tiny.onnx -e yolov7-tiny-nms.trt -p fp16

Pytorch to TensorRT another way <a href="https://colab.research.google.com/gist/AlexeyAB/fcb47ae544cf284eb24d8ad8e880d45c/yolov7trtlinaom.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> <details><summary> <b>Expand</b> </summary>

shellCopy
wget https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-tiny.pt
python export.py --weights yolov7-tiny.pt --grid --include-nms
git clone https://github.com/Linaom1214/tensorrt-python.git
python ./tensorrt-python/export.py -o yolov7-tiny.onnx -e yolov7-tiny-nms.trt -p fp16

# Or use trtexec to convert ONNX to TensorRT engine
/usr/src/tensorrt/bin/trtexec --onnx=yolov7-tiny.onnx --saveEngine=yolov7-tiny-nms.trt --fp16

</details>

Tested with: Python 3.7.13, Pytorch 1.12.0+cu113

Pose estimation

code yolov7-w6-pose.pt

See keypoint.ipynb.

<div align="center"> <a href="./"> <img src="./figure/pose.png" width="39%"/> </a> </div>

Instance segmentation (with NTU)

code yolov7-mask.pt

See instance.ipynb.

<div align="center"> <a href="./"> <img src="./figure/mask.png" width="59%"/> </a> </div>

Instance segmentation

code yolov7-seg.pt

YOLOv7 for instance segmentation (YOLOR + YOLOv5 + YOLACT)

Model	Test Size	AP<sup>box</sup>	AP<sub>50</sub><sup>box</sup>	AP<sub>75</sub><sup>box</sup>	AP<sup>mask</sup>	AP<sub>50</sub><sup>mask</sup>	AP<sub>75</sub><sup>mask</sup>
YOLOv7-seg	640	51.4%	69.4%	55.8%	41.5%	65.5%	43.7%

Anchor free detection head

code yolov7-u6.pt

YOLOv7 with decoupled TAL head (YOLOR + YOLOv5 + YOLOv6)

Model	Test Size	AP<sup>val</sup>	AP<sub>50</sub><sup>val</sup>	AP<sub>75</sub><sup>val</sup>
YOLOv7-u6	640	52.6%	69.7%	57.3%

Citation

@inproceedings{wang2023yolov7,
  title={{YOLOv7}: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors},
  author={Wang, Chien-Yao and Bochkovskiy, Alexey and Liao, Hong-Yuan Mark},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2023}
}

@article{wang2023designing,
  title={Designing Network Design Strategies Through Gradient Path Analysis},
  author={Wang, Chien-Yao and Liao, Hong-Yuan Mark and Yeh, I-Hau},
  journal={Journal of Information Science and Engineering},
  year={2023}
}

Teaser

YOLOv7-semantic & YOLOv7-panoptic & YOLOv7-caption

<div align="center"> <a href="./"> <img src="./figure/tennis.jpg" width="24%"/> </a> <a href="./"> <img src="./figure/tennis_semantic.jpg" width="24%"/> </a> <a href="./"> <img src="./figure/tennis_panoptic.png" width="24%"/> </a> <a href="./"> <img src="./figure/tennis_caption.png" width="24%"/> </a> </div>

YOLOv7-semantic & YOLOv7-detection & YOLOv7-depth (with NTUT)

<div align="center"> <a href="./"> <img src="./figure/yolov7_city.jpg" width="80%"/> </a> </div>

YOLOv7-3d-detection & YOLOv7-lidar & YOLOv7-road (with NTUT)

<div align="center"> <a href="./"> <img src="./figure/yolov7_3d.jpg" width="30%"/> </a> <a href="./"> <img src="./figure/yolov7_lidar.jpg" width="30%"/> </a> <a href="./"> <img src="./figure/yolov7_road.jpg" width="30%"/> </a> </div>

Acknowledgements

<details><summary> <b>Expand</b> </summary>

• https://github.com/AlexeyAB/darknet
• https://github.com/WongKinYiu/yolor
• https://github.com/WongKinYiu/PyTorch_YOLOv4
• https://github.com/WongKinYiu/ScaledYOLOv4
• https://github.com/Megvii-BaseDetection/YOLOX
• https://github.com/ultralytics/yolov3
• https://github.com/ultralytics/yolov5
• https://github.com/DingXiaoH/RepVGG
• https://github.com/JUGGHM/OREPA_CVPR2022
• https://github.com/TexasInstruments/edgeai-yolov5/tree/yolo-pose

</details>