sankalpvarshney / track-and-count-object-using-yolo

Advanced Vehicle Tracking and Detection System using ByteTrack, Supervision, and YOLO Algorithms

Tracking and counting of object using YOLO v8

This repository contains Python code for tracking vehicles (such as cars, buses, and bikes) as they enter and exit the road, thereby incrementing the counters for incoming and outgoing vehicles.

Installation

1. git clone https://github.com/sankalpvarshney/Track-And-Count-Object-using-YOLO.git
2. cd Track-And-Count-Object-using-YOLO
3. conda create --prefix ./env python=3.8 -y
4. conda activate ./env
5. pip install ultralytics
6. git clone https://github.com/ifzhang/ByteTrack.git
7. cd ByteTrack
8. sed -i 's/onnx==1.8.1/onnx==1.9.0/g' requirements.txt
9. pip install -q -r requirements.txt
10. python setup.py -q develop
11. pip install -q cython_bbox
12. pip install -q onemetric
13. pip install -q loguru lap
14. pip install numpy==1.22.4
15. pip install supervision==0.1.0

Usage

Firstly set the crossing line co-ordinates inside the code i.e yolov8tracker.py for the incoming and outgoing vehicles. And then execute the python code as mentioned below.

Linux

python yolov8tracker.py -i <input_video_path> -o <output_video_path>

Python

from yolov8tracker import TrackObject
obj = TrackObject(<input_video_path>,<output_video_path>)
obj.process_video()

CompressedTrafficLightObjectTracking.mp4

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

track-and-count-object-using-yolo's Issues

losing tracker_id

Hi
I'm trying to use your code on my custom data
I have a problem with an object that is going behind a wall and coming back with a different tracker_id
How to configure it that this will not happen

I tried to configure the frozen class

but it didn't help

thanks

Please provide the exact versions of the packages used

Please mention the exact versions used to develop the model/code for all the packages : like Numpy, pandas etc