<![CDATA[In the world of beauty, makeup is not only a form of self expression but also a creative skill that requires precision and an understanding of facial structures. Among all facial features, the eyes play a crucial role in defining makeup styles. Each individual has unique eye shapes such as round, monolid, upturned, almond, and downturned, which require different makeup techniques to enhance their appearance. However, many individuals struggle to identify their eye shape, leading to suboptimal makeup results. This research aims to develop an intelligent system for eye shape classification using image processing and artificial intelligence technologies. MediaPipe, a robust and lightweight framework for facial landmark detection, was employed to extract key features from the eye region, including Eye Aspect Ratio (EAR), Eye Corner (angle), and Eye Distance. A total of 1,250 images were used from various datasets including personal archives, Kaggle, and GitHub MUCT. The classification process used a Support Vector Machine (SVM) with a non-linear RBF kernel, and its performance was validated using K-Fold Cross Validation with 10 folds. The system demonstrated high accuracy for almond, downturned, monolid, and round eyes. However, classification for upturned eyes showed less optimal results, likely due to limitations in the current feature set. This study also introduces an integrated open camera interface that detects eye shape in real time and recommends suitable eye makeup styles. This research contributes to inclusive beauty technology by providing personalized makeup suggestions based on eye shape, aligning with SDG 5 (Gender Equality) and SDG 9 (Industry, Innovation, and Infrastructure). Future work will focus on improving accuracy, particularly for upturned eye classification.]]>
