Facial Diagnosis Skin Care and Makeup Recommendation Using Advanced Deep Learning Techniques

This research presents specifically an enhanced model of skin condition identification, skin color analysis, and makeup suggestion based on machine learning. The system combines a Custom Convolutional Neural Network (CNN), XGBoost, and Support Vector Machine (SVM) to help in determining skin conditions including; acne, pimples, and dark spots. The Custom CNN proves to yield higher accuracy in comparison with other models such as VGG and ResNet. Further, the skin tone, which is an important characteristic of a face mask, is formulated by paying much attention to individual facial areas or specifically the cheeks to determine the most popular color that is then compared to certain categories of skin tone. Recommendations for makeup are made using both content based and collaborative filtering approach depending on the color of the skin preference of the user. According to the results of the experimental studies, the proposed model based on the Custom CNN, XGBoost, and SVM has the high accuracy of predictions; at the same time, the use of the hybrid recommendation system allows improving the individualization of makeup recommendations. The benefit of this approach is that there is a general enhancement in the field of personalised skincare and beauty solutions.