This source describes the ONNX-based Yolo v7 object detection technique.

Items: Network weight by ONNX standard, inference code

This was from Prof. Kim Byung-Gyu on 2024. 04. 21.

Updated on 2025.03.17 by Prof. Kim Byung-Gyu

1. horse/bus.jpg : test image files
2. Inference_webcam_exam.py : we can make inference process via webcam.....!
3. Inference_exam.py : Inference process of the still image. You have to use -I option
4. yolov7-tiny.onnx : weight file for Yolov7 network model based on the onnx transformation.
5. GPU Usage: We recommend to use nVida GPU, otherwise it will be very slow.
6. How use this program?