- Object Counting and custom functions with the newest yolov7
- Crop the bounding boxes
Track_Trim.mp4
Ready to use tracking Google Colab file: https://colab.research.google.com/drive/1RK1IgQx6Fr7D2BwgCQJiUiZPclj69Myv?usp=sharing
Update 1 (9/21/2022) a new model added for stock market predicion. Model file:https://drive.google.com/file/d/1_ud8ldVz1mnRU_Z3ygXEm0kWSjgSeOEW/view?usp=sharing
- To detect: "stock_market_detect.py" file
- Example video is below
Update 2 (9/26/2022) a new file added for saving the images of bounding boxes file name: "save_bounding_boxes_images.py"
- Example photo is below
videoo.mp4
Ready-to-use Google Colab file exists. you can copy this file and run it on your colab. File: https://colab.research.google.com/drive/1Bezq6IpAraT8Ee0v8flEJco6wsLn0T3I?usp=sharing
- There are steps for applying object counter on images and videos
Features
- Count all objetcs by classes and works perfetcly on every image or on a video
- Code can run on Both (CPU & GPU)
- Video/WebCam/External Camera/IP Stream Supported
- We are going to copy offical yolov7 github page and just add "detect_and_count.py" file. That is all
- When you use Google Colab for codes, it will be way easier than working on the local computer
- clone the repository:
git clone https://github.com/WongKinYiu/yolov7
%cd yolov7
- install yolov7 model
!wget "https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7.pt"
pip install --upgrade pip
pip install -r requirements.txt
!python detect_and_count.py --weights /content/yolov7/yolov7.pt --conf 0.1 --source /content/yolov7/inference/images