Hy-Facial: Hybrid Feature Extraction by Dimensionality Reduction Methods for Enhanced Facial Expression Classification

2025-09-30

AI Summary

Facial expression classification faces challenges including difficulty in modeling high-dimensional image features and insufficient discriminability. This paper proposes Hy-Facial, the first framework to systematically integrate deep features from VGG19 with handcrafted SIFT and ORB features. To preserve multi-scale structural information, it introduces a joint dimensionality reduction strategy combining K-means pre-clustering and UMAP. Unlike PCA or t-SNE, UMAP achieves a superior balance between local neighborhood preservation and global manifold structure modeling, thereby significantly enhancing feature separability. Evaluated on the FER-2013 benchmark, Hy-Facial achieves 83.3% classification accuracy—demonstrating the synergistic advantage of hybrid feature representation and UMAP-driven dimensionality reduction. The framework establishes a novel paradigm for lightweight, efficient facial expression recognition, offering improved performance without excessive computational overhead.

Technology Category

• Machine Learning: Dimensionality Reduction/Feature Selection
• Computer Vision: Representation Learning for Vision
• Natural Language Processing: Text Classification & Sentiment Analysis

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Facial expression classification remains a challenging task due to the high dimensionality and inherent complexity of facial image data. This paper presents Hy-Facial, a hybrid feature extraction framework that integrates both deep learning and traditional image processing techniques, complemented by a systematic investigation of dimensionality reduction strategies. The proposed method fuses deep features extracted from the Visual Geometry Group 19-layer network (VGG19) with handcrafted local descriptors and the scale-invariant feature transform (SIFT) and Oriented FAST and Rotated BRIEF (ORB) algorithms, to obtain rich and diverse image representations. To mitigate feature redundancy and reduce computational complexity, we conduct a comprehensive evaluation of dimensionality reduction techniques and feature extraction. Among these, UMAP is identified as the most effective, preserving both local and global structures of the high-dimensional feature space. The Hy-Facial pipeline integrated VGG19, SIFT, and ORB for feature extraction, followed by K-means clustering and UMAP for dimensionality reduction, resulting in a classification accuracy of 83. 3% in the facial expression recognition (FER) dataset. These findings underscore the pivotal role of dimensionality reduction not only as a pre-processing step but as an essential component in improving feature quality and overall classification performance.

Problem

• Developing hybrid feature extraction combining deep learning and traditional methods
• Reducing feature dimensionality to mitigate redundancy and computational complexity
• Improving facial expression classification accuracy through optimized feature representation

Innovation

• Hybrid deep learning and traditional feature extraction
• UMAP dimensionality reduction for preserving feature structures
• Fusion of VGG19, SIFT, and ORB descriptors