|
|
--- |
|
|
title: Look-A-Like Image-Finder |
|
|
emoji: 📸🔍 |
|
|
colorFrom: green |
|
|
colorTo: gray |
|
|
sdk: streamlit |
|
|
sdk_version: 1.42.2 |
|
|
app_file: src/app/app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
<!-- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference |
|
|
|
|
|
📌 Project Overview |
|
|
|
|
|
Look-A-Like Image Finder is an AI-powered image search tool that allows users to find visually similar images based on an input image or a text query. It utilizes Pinecone for vector search, OpenAI CLIP for embeddings, and Streamlit for an intuitive user interface. |
|
|
|
|
|
✨ Features |
|
|
🔍 Image Search – Find images similar to a given input. |
|
|
📝 Text-to-Image Search – Retrieve images using natural language descriptions. |
|
|
📂 Efficient Indexing – Uses Pinecone to store and retrieve image embeddings. |
|
|
|
|
|
🛠 Tech Stack |
|
|
🖼 Model – OpenAI CLIP (Contrastive Language-Image Pretraining) |
|
|
🔍 Vector Search – Pinecone (Efficient indexing and retrieval) |
|
|
🖥 Frontend – Streamlit (Interactive web UI) |
|
|
🐍 Backend – Python |
|
|
☁ Deployment – Hugging Face Spaces |
|
|
|
|
|
📥 Resources |
|
|
🔗 Download the Dataset : {https://unsplash.com/data/lite/latest} |
|
|
🔗 Pinecone API Key Setup : {https://www.pinecone.io/} --> |
|
|
|
|
|
<!-- 🚀 How to Run the Application Locally --> |
|
|
|
|
|
<!-- Step 1: Clone the Repository --> |
|
|
<!-- Copy and paste the following command into your terminal --> |
|
|
<!-- git clone https://github.com/Vela-Test1993/lookalike-image-finder.git --> |
|
|
<!-- cd lookalike-image-finder --> |
|
|
|
|
|
<!-- Step 2: Install Dependencies --> |
|
|
<!-- Run the following command to install the required packages --> |
|
|
<!-- pip install -r requirements.txt --> |
|
|
|
|
|
<!-- Step 3: Start the Application --> |
|
|
<!-- Use the command below to launch the app --> |
|
|
<!-- streamlit run src/app/app.py --> |
|
|
|
|
|
<!-- Step 4: Open in Your Browser --> |
|
|
<!-- The application will automatically open in your default web browser. --> |
|
|
|
|
|
<!-- ⚠ Important: API Key Setup --> |
|
|
<!-- Create a .env file in the project's root directory and store your Pinecone API key inside it. --> |
|
|
|
|
|
|
|
|
<!-- 🔑 Setting Up the Application Components --> |
|
|
|
|
|
<!-- 🟢 Pinecone: Vector Database --> |
|
|
<!-- 1️⃣ Log in to your Pinecone account. --> |
|
|
<!-- 2️⃣ Retrieve your Pinecone API key. --> |
|
|
<!-- 3️⃣ Store the API key securely in the .env file. --> |
|
|
|
|
|
<!-- 📂 Dataset: Image Processing --> |
|
|
<!-- 1️⃣ Download the dataset from the following link: https://unsplash.com/data/lite/latest --> |
|
|
<!-- 2️⃣ Convert the images into vector embeddings using OpenAI CLIP. --> |
|
|
<!-- 3️⃣ Store the embeddings in Pinecone for efficient retrieval. --> |
|
|
|
|
|
<!-- 🎨 Streamlit: Web Interface --> |
|
|
<!-- 1️⃣ Streamlit is used to build the user-friendly UI/UX for the application. --> |
|
|
<!-- 2️⃣ The frontend allows users to search for similar images using image or text queries. --> |
|
|
<!-- 3️⃣ The interface is interactive and easy to navigate. --> |
|
|
|
|
|
|
