Makup Recomandation System

The Makeup Recommendation System is a web-based application that recommends personalized skincare and makeup products based on the user's skin metrics, inferred from a selfie using computer vision algorithms. By analyzing skin tone, skin type, and acne concerns, this system suggests the most suitable products for each user, enhancing the shopping experience. The web application is built using React for the frontend and Flask for the backend, utilizing Image Processing and Convolutional Neural Networks (CNN) for accurate skin analysis.

Project Features:

• Selfie Analysis: The application uses face detection technology to capture a selfie, ensuring the image meets the necessary criteria for accurate skin analysis, such as proper luminance and the face being the dominant part of the image.
• Skin Metrics Extraction: Using advanced image processing techniques and CNN models, the system detects the skin tone, type (dry, oily, normal), and acne concern level from the selfie.
• Product Recommendation: The system recommends skincare and makeup products by calculating the cosine similarity between the user's skin metrics and the product attributes, ensuring the suggestions are highly relevant to the user's needs.
• Real-time User Interaction: The user can take a selfie, view their skin metrics, and adjust them before receiving personalized product recommendations.
• Frontend Routes: The web application features various routes, including the selfie capture page, the results page for skin metrics, and the product recommendation page where users can view detailed product information.

Applications of the Makeup Recommendation System:

• Personalized Skincare: Users can get skincare and makeup product suggestions tailored to their unique skin features, improving their skincare routine and makeup choices.
• Beauty E-commerce: Integrated into e-commerce platforms, this system can enhance the shopping experience by offering relevant product recommendations, increasing customer satisfaction and conversion rates.
• Health and Wellness: This system can also be used to recommend products that address specific skin concerns such as acne, dryness, or redness, promoting healthier skin.

Technology Stack:

• Frontend: React.js for dynamic user interfaces and smooth interactions.
• Backend: Flask for handling API requests and image processing tasks.
• Image Processing: face-api.js for face detection, and CNN models for skin tone and type analysis.
• Cosine Similarity: Used for recommending the most relevant products based on user skin metrics.

This application provides a unique way to offer personalized beauty product recommendations by utilizing advanced technologies such as computer vision and machine learning.