Object Finder

Overview

Object Finder is a Python project that uses the DeepLabV3+ model to identify people in an image and highlight them with a colored overlay. The project features a user interface (UI) created with PyQt5 for an interactive experience.

Requirements

• Python 3.x

• Install the required libraries using the following command:

  pip install -r requirements.txt

Usage

1. Clone the repository:

   git clone /~https://github.com/your-username/ObjectFinder.git
   cd ObjectFinder

2. Run the project using the provided script:

   python3 ./src/main.py

Details

• The project uses the DeepLabV3+ model from torchvision to perform semantic segmentation on the input image.
• Identified people are highlighted with a specified color and transparency level.
• OpenCV is utilized for image processing.
• The UI is built with PyQt5, providing a user-friendly interface for image manipulation.

UI Screenshots

Caption: The main interface of the Object Finder application.

Caption: Selecting an image using the UI.

File Structure

• src/main.py: The main script to run the object finder.
• data/sample_images/: Contains sample images for testing.
• requirements.txt: Lists the required Python libraries and their versions.

Contributions

• @Diogo Faria Martins
• @김서해
• @Bastien Boymond
• @Hugo Nini

Acknowledgments

• DeepLabV3+ model: torchvision.models.segmentation.deeplabv3_resnet101
• OpenCV: opencv-python

Note: This project adheres to specific guidelines. Numpy is intentionally excluded, and only input/output functions from OpenCV are used.

Feel free to customize the project to fit your specific needs!