https://github.com/dlod-openvino/yolov5_infer

Do the YOLOv5 model inference by OpenCV/OpenVINO based on onnx model format
https://github.com/dlod-openvino/yolov5_infer

Last synced: 3 months ago
JSON representation

Do the YOLOv5 model inference by OpenCV/OpenVINO based on onnx model format

• Host: GitHub
• URL: https://github.com/dlod-openvino/yolov5_infer
• Owner: dlod-openvino
• License: mit
• Created: 2022-03-19T08:14:00.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2023-02-06T02:16:34.000Z (about 2 years ago)
• Last Synced: 2024-08-02T01:20:03.265Z (7 months ago)
• Language: Python
• Size: 77.1 KB
• Stars: 13
• Watchers: 2
• Forks: 3
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-yolo-object-detection - dlod-openvino/yolov5_infer - openvino/yolov5_infer?style=social"/> : Do the YOLOv5 model inference by OpenCV/OpenVINO based on onnx model format. (Lighter and Deployment Frameworks)
• awesome-yolo-object-detection - dlod-openvino/yolov5_infer - openvino/yolov5_infer?style=social"/> : Do the YOLOv5 model inference by OpenCV/OpenVINO based on onnx model format. (Lighter and Deployment Frameworks)

README

        
# YOLOv5 Inference Demo by OpenCV and OpenVINO

Do the YOLOv5 model inference by OpenCV/OpenVINO based on onnx model format

### Installation

Clone the repository

$ git clone https://github.com/ultralytics/yolov5.git

Enter the repository root directory

$ cd yolov5

Install the required packages from your cloned repository root directory

$ pip install -r requirements.txt

### Export the YOLOv5 model to onnx model

$ python export.py --weights yolov5s.pt --include onnx

### demo code

+ infer_by_opencv.py: do the inference by the OpenCV DNN module

+ infer_by_openvino.py: do the inference by the OpenVINO<=2021.4.2

+ infer_by_openvino2022.py: do the inference by the OpenVINO>=2022.1

+ test_env.cpp: test the VS2019 development environment based on OpenVINO>=2022.1 and OpenCV, C++ version

+ yolov5_ov2022_cam.cpp: sample code about do the yolov5 inference by USB camera. OpenVINO>=2022.1 C++ version

+ yolov5_ov2022_image.cpp:sample code about do the yolov5 inference on one image. OpenVINO>=2022.1 C++ version

+ infer_with_openvino_preprocess.py: sample code about do the yolov5 inference in ASYNC mode with OpenVINO preprocessing API. OpenVINO>=2022.1 Python version

+ infer_with_openvino_preprocess_sync.py: sample code about do the yolov5 inference in SYNC mode with OpenVINO preprocessing API. OpenVINO>=2022.1 Python version

+ preprocessing_with_saving_to_IR.py: sample code about export the IR model with preprocessing

+ openvino2022-device-for-mqtt.py: push the inference result of OpenVINO>=2022.1 to EdgeX by MQTT

+ ov_cvmart_sample.ipynb: cvmart newbie task OpenVINO>=2022.1 sample [基于YOLOv5的新手任务](https://www.cvmart.net/document)

+ yolov5_ov2022_sync_dGPU.py: do the yolov5 Sync inference on intel discreate GPU by OpenVINO2022.2, pls "pip install -U yolort" firstly, refer to https://github.com/zhiqwang/yolov5-rt-stack

+ yolov5_ov2022_async_dGPU.py: do the yolov5 Async inference on intel discreate GPU by OpenVINO2022.2, pls "pip install -U yolort" firstly, refer to https://github.com/zhiqwang/yolov5-rt-stack

+ ji.py: cvmart newbie task, auto test script.[基于YOLOv5的新手任务,编写测试脚本](https://www.cvmart.net/document)

+ yolov5seg_ov2022_sync_dGPU.py: [YOLOv5-Seg instance segmentation Python sample code by >=OpenVINO2022.2 on Intel A770](https://mp.weixin.qq.com/s/K3wP5YLAU4p5jsdiMYjuMg)

+ yolov5_async_infer_queue.py: [YOLOv5 Async Infer based on OpenVINO AsyncInferQueue](https://mp.weixin.qq.com/s/sHRC09l5g2YRYhJzzMNI-Q)

+ yolov5_openvino_sync_dGPU.cpp: [YOLOv5 Sync Infer C++ Demo on intel discreate GPU(A770)](https://mp.weixin.qq.com/s/S_p5kN-M0xDsCzi-5Qbo0w)

+ yolov5seg_openvino_dGPU.cpp: YOLOv5-Seg Model Sync Infer C++ Demo on intel discreate GPU(A770)

### references

+ https://docs.ultralytics.com/quick-start/

+ https://github.com/ultralytics/yolov5/releases/tag/v6.1

+ https://docs.openvino.ai/latest/notebooks/002-openvino-api-with-output.html#

+ https://towardsdatascience.com/yolo-v4-or-yolo-v5-or-pp-yolo-dad8e40f7109