Real-time object detection application

Web application for real-time object detection on video streaming via web browser.

Watch the demo video.

Installation

Create and activate an virtual environment, as follows:

$ cd cloned/directory/
$ python -m venv env
$ env/Scripts/activate

After have installed and activated the environment, install all the dependencies:

$ pip install -r requirements.txt

After that, you can run the following command and access the application at 127.0.0.1:5000 on your browser.

$ python application.py

obs.: This application was tested only on Google Chrome.

Download model

To download the yolov3.weights, just run:

$ cd models/
$ python dl-weights.py

Usage

• Camera preview: Enables and disables the webcam preview.
• Run detection model: Enables and disables the detection model.
• Exposure: Buttons which increase or decrease camera exposure stops by 1.
• Contrast: Buttons which increase or decrease camera contrast stops by 4.
• Reset camera: Reset all camera settings based on camera_settings.log file.