Image Comparer

A small Python utility for comparing and inspecting image sets. This project helps you compare images across folders, run quick visual checks, and (optionally) compute similarity using a pretrained model.

Purpose

• Organize and compare image sets (examples kept under Images/).
• Provide a lightweight script (app.py) to run comparisons or visual inspections.
• Include a pretrained ResNet checkpoint under pretrained_models/ for model-based similarity or feature extraction.

Requirements

• Python 3.8+
• See requirements.txt for exact Python package dependencies.

Installation

1. Create and activate a Python virtual environment (recommended):

   python -m venv .venv
   .\\.venv\\Scripts\\Activate.ps1

2. Install dependencies:

   pip install -r requirements.txt

Quick Usage

• The main script is app.py. It runs the project's comparison logic. Basic usage:

  python app.py

• By default the repository has an Images/ folder with example subfolders. You can point app.py at whichever folders you want to compare, or edit the script to change paths and options.

• If the project uses the pretrained model, confirm the checkpoint at pretrained_models/checkpoints/ is present (for example resnet50_fbgemm-23753f79.pth). The code will load it as needed for feature extraction.

Project Layout

• app.py — main entrypoint for running comparisons
• requirements.txt — Python dependencies
• Images/ — sample image folders (A, B, Base Images, Solution Images)
• pretrained_models/ — model checkpoints used for feature extraction