https://github.com/we0091234/yolov7-face-tensorrt

yolov7-face tensorrt
https://github.com/we0091234/yolov7-face-tensorrt

face-detection face-landmark-detection yolov5-face yolov5-face-landmark yolov7 yolov7-face

Last synced: 3 months ago
JSON representation

yolov7-face tensorrt

• Host: GitHub
• URL: https://github.com/we0091234/yolov7-face-tensorrt
• Owner: we0091234
• Created: 2022-12-10T09:37:23.000Z (about 2 years ago)
• Default Branch: master
• Last Pushed: 2023-02-24T03:45:40.000Z (almost 2 years ago)
• Last Synced: 2024-11-09T16:44:12.938Z (3 months ago)
• Topics: face-detection, face-landmark-detection, yolov5-face, yolov5-face-landmark, yolov7, yolov7-face
• Language: C++
• Homepage:
• Size: 20.3 MB
• Stars: 43
• Watchers: 1
• Forks: 7
• Open Issues: 7
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-yolo-object-detection - we0091234/yolov7-face-tensorrt - face-tensorrt?style=social"/> : yolov7-face TensorRT. (Applications)
• awesome-yolo-object-detection - we0091234/yolov7-face-tensorrt - face-tensorrt?style=social"/> : yolov7-face TensorRT. (Applications)

README

        
# yolov7-face TensorRT

## yolov7-face

  The Pytorch implementation is [yolov7-face](https://github.com/derronqi/yolov7-face).

## **onnx export**

  step1.  git clone  https://github.com/derronqi/yolov7-face

  step2.  modify models/export.py

```

 #line 72

 #output_names = None

 # modified into:

 output_names = ["output"]

```

   step3.  modify models/yolo.py   class  IKeypoint

```

#line 308

# return x if self.training else (torch.cat(z, 1), x)

# modified into:

return x if self.training else torch.cat(z, 1)

```

  step4. Export to onnx model

```

 cd yolov7-face

 python models/export.py --weights yolov7s-face.pt --grid 

```

## How to Run, yolov7s-face  as example

1. Modify the tensorrt cuda opencv path in CMakeLists.txt

   ```

   #cuda 

   include_directories(/mnt/Gu/softWare/cuda-11.0/targets/x86_64-linux/include)

   link_directories(/mnt/Gu/softWare/cuda-11.0/targets/x86_64-linux/lib)

   #tensorrt 

   include_directories(/mnt/Gpan/tensorRT/TensorRT-8.2.0.6/include/)

   link_directories(/mnt/Gpan/tensorRT/TensorRT-8.2.0.6/lib/)

   ```

2. build

   ```

   1. mkdir build

   2. cd build

   3. cmake ..

   4. make

   ```

3. onnx  to tensorrt model

   ```

   ./onnx2trt/onnx2trt  ../onnx_model/yolov7s-face.onnx ./yolov7s-face.trt  1

   ```

4. inference

   ```

   ./yolov7_face yolov7s-face.trt ../images

   ```

   The results are saved in the build folder.

   ![image](result/test.jpg)

## contact

Tencent qq group:  871797331