A Hybrid AI Hairstyle Recommendation System Based on Face Shape Classification

View at publisher Summarise with VIRAT
Authors: Muhammad Amirul Ilham Yusri, Mohd Talmizie Amron, Zawawi Abdul Wahab
Cited by: 0
References: 18

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper presents a hybrid AI hairstyle recommendation system that combines automated face shape classification with generative recommendations and curated video tutorials. The system integrates YOLOv8 for detecting and classifying face shapes, the Gemini API for generating personalised hairstyle suggestions, and the YouTube Data API for retrieving relevant tutorial content, all deployed within a Flutter-based mobile application. A dataset of 297 labelled facial images was used to train a three-class face shape model (oval, round, square). The proposed system achieved moderate classification performance, sufficient to support a proof-of-concept pipeline. It was evaluated for user experience using the System Usability Scale (SUS), yielding a score of 83.36, indicating high usability. The contribution of this work is primarily integrative and deployment-oriented: it demonstrates how existing AI components can be orchestrated into an end-to-end personalised recommendation experience. Limitations related to dataset size, model performance, population coverage, and ethical considerations are discussed, along with directions for expanding the dataset, improving accuracy, incorporating richer feedback, and mitigating bias and privacy risks.

Fields of Study & Subjects

Recommender Systems and Techniques Face recognition and analysis Advanced Neural Network Applications

Publication details

Journal / source
International Conference on Cognitive Systems and Computer Interaction (ICoSCI)
Pages
1–6
Publication date
15 January 2026
Language
EN