A Novel Approach for Smart Skincare Solution: Derma AI Recommendation System
B N Ramva, Amrutha Surabhi, Ananya V Adiga, Priya S Kulkarni, Sinchana Gajanan Hegde · 2025
This paper proposes a novel automatic skin type detection and personalized makeup product recommendation system. Transfer learning with Mobile Net architecture is utilized to classify face images into oily, dry, normal, and acne-prone skin types. The classification output is further integrated with a recommendation system which recommends suitable skin care based on the identified skin type. The web application is built using Flask, resulting in a clean and user-friendly interface. The classification output is also combined with a recommendation system that proposes appropriate makeup products depending on the identified skin type. Unlike existing literature, the proposed approach unifies both classification and recommendation features, enabling better personalization. By incorporating datasets with diverse skin types, the system achieves strong classification performance and provides customized cosmetic recommendations. This work is a contribution to the emerging personalized beauty technology through the fusion of computer vision and recommendation systems for improving consumer cosmetic choice experiences.