International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 04 | Apr 2025
p-ISSN: 2395-0072
www.irjet.net
AI-Powered Face Shape Detection for Personalized Hairstyle Recommendations Using Deep Learning Models. Nikita Shinde1, Sanjeevani Tagade2, Saloni Varma3 123Bachelor of Technology in Department of Data Science, Usha Mittal Institute of Technology, Mumbai, India
---------------------------------------------------------------------***--------------------------------------------------------------------utilizing deep learning models, such as CNN-based Abstract - This research introduces a novel method to face
architectures like VGG16 and MobileNet, to enhance face shape classification. By training the system on a varied collection of facial images, we aim to develop a model capable of consistently recognizing different face shapes. Furthermore, we concentrate on creating an easy-to-use interface that enables users to effortlessly upload their photos and receive hairstyle recommendations tailored to their facial features. Ultimately, this research aims to connect technology with personal styling, simplifying the process of hairstyle selection.
shape detection in order to better recommend hairstyles. Precise recognition of the shape of the face__ whether oval, round, heart, square or oblong is essential in recommending hair styles that match as nearly as possible an individual's personal characteristics. Our method combines advanced image processing and machine learning algorithms with AI__ integrating deep learning models that have been pre-trained, such as a CNN-based architecture like VGG16 or MobileNet. It identified facial landmarks using deep learning models trained on a comprehensive data set of face images. In addition, to combat variations in facial expressions and lighting conditions, we have extended the system with artificial intelligence techniques that help it better handle these different circumstances. Empirical results demonstrate that when our face shape detection model is integrated with a hair do counseling system, the system offers hair suggestions tailored according to face shape as determined by the headband. Methodology efficiency and accuracy demonstrated by experiment will directly indicate the potential application prospects for personal grooming mode plus Virtual Styling Tools.
2. LITERATURE SURVEY In this it involves segmenting the head and identifying the face plane to minimize shadows. It uses an algorithm that treats 3D body points as evaluation features, employing Eigenvectors for refinement and techniques like Ellipsoid Fitting and Mahalanobis distance for segmentation. The HaarCascade Classifier and dlib functions are used for realdata detection, improving classification accuracy[1]. Face recognition has advanced over 30 years, widely used in commercial and law enforcement sectors. Despite progress, challenges like lighting and angle variations remain. Face shape detection benefits industries like fashion and ecommerce. CNNs are highly effective for face classification, automatically identifying features with minimal training. Key challenges include adapting to varied conditions, improving real-time processing, and enhancing personalization. Future improvements focus on robustness, real-time capabilities, and integrating multimodal data for more accurate recognition[2]. Recent advances in face shape classification, mainly using CNNs, have improved accuracy in fashion and healthcare. Challenges like image quality and expressions remain, requiring stronger algorithms and real-time improvements[3]. Advancements in hairstyle recommendation use CNNs to analyze face shapes and suggest suitable styles. Real-time simulations improve user experience, but challenges like lighting, hair diversity, and processing speed remain, requiring better datasets and personalization [4]. Hybrid models combining CNNs with other techniques improve hairstyle recommendations. Challenges like lighting, expressions, and hair types persist. Augmented reality and real-time simulations enhance user experience, driving personalized, data-driven advancements [5]. The CelebA dataset, known for its diverse celebrity facial images with annotations, has been adapted for hairstyle recommendation. Researchers use it to train CNN models,
Key Words: Face shape detection, hairstyle recommendation, deep learning, convolutional neural networks (CNN), VGG16, MobileNet, image processing, facial landmarks, machine learning, virtual styling, personal grooming.
1.INTRODUCTION Choosing an appropriate hairstyle in the current era of personal care and style can be daunting, given the multitude of choices available. Many people find it challenging to identify a look that suits their particular facial structure, often resorting to experimentation or seeking expert guidance, which can be both time-consuming and potentially unsuccessful. This research investigates the creation of an artificial intelligence-driven system that identifies face shapes and suggests tailored hairstyles to address this issue. Conventional approaches to hairstyle selection often lack accuracy, as they rely on subjective judgments or generic recommendations. Although some digital applications claim to offer hairstyle advice, many fail to precisely evaluate face shapes or provide truly personalized suggestions. This underscores the necessity for a smart, user-friendly solution that employs cutting-edge technology to accurately analyze facial characteristics. Our study examines the potential of
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 539