README.md

metadata

title: Image Object Detection for Distributor Salesmen’s Customer Visits
emoji: 📸
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.23.1
app_file: app.py
pinned: false
license: mit

Project Description

This project aims to develop a computer vision-based solution to streamline the image verification process for FMCG distributor admins. Salesmen visiting customers are required to capture and submit 5-10 photos per customer as proof of visit, often including selfies, product displays, competitor surveys, and promotional materials. With each salesman visiting 25-40 customers daily, admins face the tedious task of manually reviewing up to 400 images per salesman every day. By implementing an image object detection system, this solution will automatically validate submitted images against admin-defined criteria, such as ensuring the presence of a face in selfies or verifying product displays. This automation reduces administrative workload, enhances efficiency, and ensures compliance with visit requirements.

Prerequisites

Before you begin, ensure you have met the following requirements:

1. Flutter Setup
   • On your device, install Flutter SDK and set up the environment.
2. Python Setup
   • Python 3.8 or later
   • pip (Python package installer)

Running the Project

Follow these steps:

1. Set Up Backend Server

   cd backend
   
   # Create virtual environment
   python -m venv .venv
   
   # Activate virtual environment
   # For macOS/Linux:
   source .venv/bin/activate
   
   # For Windows:
   .\venv\Scripts\activate
   
   # Install dependencies
   pip install -r requirements.txt
   
   # Start the server
   python -m clip_server config.yml
   

2. Verify Server Status Visit http://0.0.0.0:51000/status in your browser.

3. Access & download frontend from https://github.com/Psianturi/Final-Project-AI-GPT-Bootcamp-Q4-2024-Encode-Club/tree/main/frontend/image_detection_app

4. Open frontend folder & run with android studio or visual studio code

5. Use flutter run command & run on emulator/device

Report / Documentation

Salesmen visiting customers are required to capture and submit photos per customer as proof of visit, often including selfies, product displays, competitor surveys, and promotional materials.

System will automatically validate submitted images against admin-defined criteria, such as ensuring the presence of a face in selfies or verifying product displays. When the image fulfills the criteria, the system will return the result as shown below (Target Achieved).

When criteria are not met, the system will return the result as shown below (Target Not Achieved).

Demo Video

visit this link: Demo Video