In recent years, the demand for intelligent skincare solutions has grown significantly, driven by increasing awareness of dermatological health and personalized cosmetic needs. This paper presents a novel AI-powered framework for real-time detection of skin anomalies using the YOLO (You Only Look Once) object detection algorithm, integrated with a personalized product recommendation engine. The system is designed to identify common skin issues such as acne, pigmentation, and dryness by processing facial images and extracting anomaly-specific features. Leveraging the efficiency of YOLOv8 for fast and accurate detection, the model ensures high precision in identifying multiple skin conditions under varying lighting and skin tone conditions. Following the detection phase, an AI-based recommendation engine analyzes the detected conditions and user-specific parameters to suggest tailored skincare products from a curated database. The recommendation logic incorporates skin type, severity of anomalies, and product efficacy history to enhance personalization. Experimental results demonstrate promising detection accuracy and improved user satisfaction in recommendation relevance compared to traditional systems. The integration of real-time detection with intelligent recommendation marks a significant advancement in AI-driven dermatological assistance, enabling a scalable solution for consumer-level skincare monitoring and advisory. This framework paves the way for future enhancements involving dynamic learning from user feedback and integration with teledermatology platforms.

Keywords: Skin Anomaly Detection, YOLO, Deep Learning, AI Recommendation Engine, Personalized Skincare, Computer Vision, Real-Time Analysis, Dermatological AI