https://github.com/nvinayvarma189/object-counting-video

https://github.com/nvinayvarma189/object-counting-video

Last synced: 6 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/nvinayvarma189/object-counting-video
• Owner: nvinayvarma189
• License: mit
• Created: 2021-06-24T08:56:14.000Z (over 4 years ago)
• Default Branch: main
• Last Pushed: 2021-06-26T08:32:59.000Z (over 4 years ago)
• Last Synced: 2025-02-08T06:12:14.667Z (8 months ago)
• Language: Python
• Size: 40 KB
• Stars: 1
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          

# Object Counting Video

  
Table of Contents

  


    


      About The Project

      


        
Built With

      

    

    


      Getting Started

      


        
Prerequisites

        
Installation

      

    

    
Usage

    
Contributing

    
License

    
References

  


## About The Project

  A simple program to calculate the cumulative count of instances of one or more objects present in a video. Confidence thresholds, line placement, objects list, etc are customizable to make it as generic as possible.

  

![counting_input](https://user-images.githubusercontent.com/30727381/123505490-0189db80-d67d-11eb-9c0f-9077c470fa08.png)

![counting_output](https://user-images.githubusercontent.com/30727381/123505493-051d6280-d67d-11eb-94d5-cce437d10b8b.png)

### Built With

* [YoloV3](https://arxiv.org/abs/1804.02767)

* [OpenCV](https://opencv.org/)

* [Python3](https://www.python.org/)

## Getting Started

To get a local copy up and running:

### Prerequisites

1. `git clone https://github.com/nvinayvarma189/object-counting-video`

2. `cd object-counting-video && mkdir input/ && mkdir output/` 

3. Downlaod a traffic video clip such as [this](https://www.youtube.com/watch?v=jjlBnrzSGjc) and place it unde `input` folder under your `object-counting-video` folder.

4. Downlaod [Yolov3 weights](https://pjreddie.com/media/files/yolov3.weights) file and place in under [`yolo-coco`](https://github.com/nvinayvarma189/object-counting-video/tree/main/yolo-coco) folder.

### Installation

1. `python -m venv env`

2. `source env/bin/activate`

3. `pip install -r requirements.txt`

### Usage

1. If you have the above-mentioned prerequisites and have followed the installation steps then you can run this command 

      - `python main.py count_objects_in_video`

2. If you want to pass any arguments to the script, you can do pass them like this

      - `python main.py count_objects_in_video --input_video_path='/some/custom/path`

      - `python main.py count_objects_in_video --line_coords='[(x1, y1,), (x2, y2)]'`

      - `python main.py count_objects_in_video --output_json_path='/some/custom/path/output.json`

      - `python main.py count_objects_in_video --objects_to_count='["object1", "object2", "object3"]`

3. To list all possible arguments along with their default values:

      - `python main.py -- --help`

### Contributing

Coming Soon...

### License

This repository is licensed with the [MIT License](https://github.com/nvinayvarma189/object-counting-video/blob/main/LICENSE)

### References

- [YoloV3 paper](https://arxiv.org/abs/1804.02767) and [Pretrained YoloV3 weights](https://pjreddie.com/media/files/yolov3.weights)

- [Object Detection and Tracking blog post](https://towardsdatascience.com/object-detection-and-tracking-in-pytorch-b3cf1a696a98)

- [SORT: A Simple, Online and Realtime Tracker](https://github.com/abewley/sort/blob/master/sort.py)

- [Object counting using Yolo and Sort](https://github.com/HodenX/python-traffic-counter-with-yolo-and-sort)