Fact-checked by Grok 2 weeks ago

Eigenface

An eigenface is a principal component of the set of images depicting human faces, derived through (PCA) of the formed by treating face images as high-dimensional vectors; this technique forms the basis of an early and influential method for automated face recognition in . Developed by Matthew Turk and in 1991, eigenfaces enable the efficient representation and comparison of faces by projecting them into a low-dimensional "face space" spanned by the most significant eigenvectors, which capture the primary variations in facial appearance across a training dataset. The method begins with a collection of training face images, which are normalized for (e.g., positioning eyes and mouth) and , then subtracted by the face to center the data. The of these centered images is computed, and its eigenvectors—termed eigenfaces—are ordered by their corresponding eigenvalues to select the top M components that account for the majority of variance. A novel input face is then projected onto these eigenfaces to obtain a set of weights, representing the face as a point in the reduced ; recognition occurs by measuring the between this projection and those of known individuals, with a to detect non-faces. This approach, building on earlier work by Sirovich and Kirby (1987) for image representation, achieves near-real-time performance and high accuracy (e.g., 96% under controlled variations) in constrained environments like frontal views with consistent illumination. Eigenfaces marked a pivotal advancement in facial recognition by providing a holistic, appearance-based alternative to feature-specific or techniques, serving as a for subsequent algorithms such as Fisherfaces and kernel variants. Despite its computational efficiency and role in early commercial systems (e.g., licensed to companies such as Viisage Technology following its patenting in ), the method's reliance on global variance makes it vulnerable to variations in pose, expression, scale, , and especially lighting, which can dominate the principal components and degrade performance outside ideal conditions. Its foundational impact persists in modern research, influencing the evolution toward deep learning-based systems while highlighting the challenges of robust biometric identification.

Fundamentals

Definition and Overview

Eigenfaces are a set of vectors derived from the principal components of variation in a collection of face images, serving as characteristic features for representing facial patterns. These basis vectors, known as eigenfaces, emerge from applying to the of centered face images, capturing the most significant directions of variability across the . Visually, eigenfaces manifest as ghostly, averaged face-like patterns, where each eigenvector is displayed as an image highlighting the relative contributions of locations to the overall variation. These patterns often resemble ethereal outlines of faces, emphasizing global structural elements rather than fine details, and they inherently encode major sources of difference such as pose and lighting across the training images. The core purpose of eigenfaces lies in , transforming high-dimensional face images into a compact "face " spanned by these basis vectors, which facilitates efficient storage, comparison, and of . By projecting images onto this lower-dimensional , eigenfaces enable systems to focus on essential features while discarding noise and minor variations. In a basic workflow, eigenfaces are generated by training on a dataset of face images to compute the principal components, after which new face images are projected onto the resulting face space to yield coordinate representations for subsequent analysis.

Principal Component Analysis in Images

(PCA) is a statistical used to identify patterns in high-dimensional data by transforming it into a lower-dimensional space while maximizing the variance captured by the new coordinates, known as principal components. These components are orthogonal directions in the data that successively account for the largest amounts of variability, enabling that preserves essential information and facilitates analysis or . When applied to images, such as face photographs, PCA requires representing each as a point in a high-dimensional . For instance, a typical 256 by 256 is flattened into a one-dimensional of 65,536, where each element corresponds to a value, allowing the set of images to form a in this expansive space. To apply , the data is first centered by subtracting the mean image from each image , resulting in a mean-subtracted data matrix \Phi of size d \times N, where d is the image dimension (e.g., ) and N is the number of images. The \Sigma of this centered data is then computed as \Sigma = \frac{1}{N} \Phi \Phi^T, which captures the variance-covariance structure across the pixel dimensions. The eigen of \Sigma yields eigenvectors v_i and eigenvalues \lambda_i, satisfying \Sigma v_i = \lambda_i v_i for i = [1](/page/1), \dots, d. Here, the eigenvectors v_i represent the principal components (termed eigenfaces in the context of face images), and the corresponding eigenvalues \lambda_i quantify the amount of variance explained by each component, with larger \lambda_i indicating greater importance. Principal components are selected by sorting the eigenvectors in descending order of their eigenvalues and retaining the top k such that they capture a substantial portion of the total variance, often determined by a threshold on the cumulative explained variance ratio \sum_{i=1}^k \lambda_i / \sum_{i=1}^d \lambda_i. For example, components are chosen to retain at least 95% of the total variance, balancing dimensionality reduction with information preservation.

Historical Development

Origins and Key Publications

The eigenfaces technique originated in the late 1980s at the MIT Media Laboratory, where researchers Matthew Turk and developed an appearance-based method for face recognition by adapting (PCA) from earlier applications to the domain of human faces. This work built upon foundational studies, such as those by Sirovich and Kirby, who in 1987 demonstrated PCA's utility for low-dimensional representation of face images under controlled conditions. The seminal publication introducing eigenfaces as a practical recognition tool was the 1991 paper "Eigenfaces for Recognition" by Turk and Pentland, published in the Journal of Cognitive Neuroscience. In this work, they presented a system that encodes face images as vectors in a high-dimensional space and uses PCA to derive a set of orthogonal basis images—termed eigenfaces—that capture the principal variations among faces, enabling efficient comparison and identification. The method was designed for near-real-time performance, addressing challenges in automated face tracking and recognition within computer vision. Early experiments in the paper utilized a custom database of over 2,500 face images from 16 , captured under varying conditions including three lighting directions, three head sizes, and three orientations to simulate real-world variability. These tests demonstrated the technique's robustness, achieving high accuracy by projecting novel faces onto the subspace and measuring distances to known , thus establishing eigenfaces as a foundational in face research.

Evolution and Impact

Following the seminal 1991 work by and Pentland introducing eigenfaces as a principal component analysis-based approach to face recognition, the method rapidly expanded in the early 1990s through integrations into practical systems. In the late 1990s and early 2000s, eigenfaces influenced commercial biometric security applications, such as user authentication and physical systems. This integration facilitated near-real-time processing, enabling deployments in security and . Academically, eigenfaces exerted profound influence on and , with the original paper amassing over 21,500 citations as of 2025, underscoring its role in popularizing subspace learning methods. The approach established as a foundational tool for in image analysis, inspiring subsequent algorithms like Fisherfaces and modular eigenspaces that addressed limitations in holistic representations. Its impact extended to , where -based eigenface techniques became a staple in introductory curricula, providing an accessible entry point for teaching feature extraction and pattern recognition concepts. Key milestones in the evolution included 1994 extensions by Pentland and colleagues, who developed view-based and modular eigenspaces to enable pose-invariant by decomposing faces into component-specific subspaces, such as eyes and nose, rather than treating the entire image holistically. This advancement improved robustness to viewpoint variations and influenced benchmarking practices, notably through the FERET dataset introduced in 1996, where eigenfaces served as a primary for evaluating face algorithms across thousands of images under controlled conditions. These developments solidified eigenfaces' legacy in shaping standardized protocols for the field.

Computation Methods

Data Preparation and Training

The preparation of data for eigenface generation begins with assembling a consisting of face images captured under controlled conditions to minimize extraneous variations. These images are typically aligned and cropped to focus on the facial region, with a standard resolution such as 92×112 pixels to ensure uniformity across the set. In seminal implementations, datasets like the () database are employed, featuring 400 images from 40 subjects, each with 10 variations in pose and lighting. For robust eigenfaces, training sets generally range from 100 to 500 images across 20 to 50 subjects, providing sufficient diversity while remaining computationally feasible. Preprocessing steps are essential to standardize the images and address common challenges in face data. Initial and cropping isolate the face from background elements, often using techniques like elliptical masking to exclude non-facial areas. Images are converted to to simplify representation and reduce color-induced noise, resulting in vectors suitable for analysis. handles variations in scale, orientation, and illumination; this may involve resizing to fixed dimensions, rotating for alignment based on facial landmarks (e.g., eyes and ), and applying to mitigate lighting differences. These steps ensure the dataset captures intrinsic facial structure rather than environmental artifacts. A key aspect of preparation is centering the data around the mean face to highlight deviations that define individual characteristics. The mean face \Psi is calculated as the average of all training images \{\tau_i\}, where i indexes the images in the set. Each image is then centered by subtracting this mean: \phi_i = \tau_i - \Psi, producing a mean-subtracted set \{\phi_i\} that facilitates subsequent statistical analysis. This centering step, performed after initial normalization, aligns the data distribution for effective principal component extraction.

Eigenvector Extraction and SVD Connection

The computation of eigenfaces begins with the formation of the from the prepared, mean-subtracted images. Let \{\phi_i\}_{i=1}^N denote the vectors representing these centered images, each of d (where d is the total number of pixels). The sample is then given by \Sigma = \frac{1}{N} \sum_{i=1}^N \phi_i \phi_i^T, which is a d \times d symmetric positive semi-definite capturing the variance structure across the set. Direct extraction of the eigenvectors of \Sigma yields the eigenfaces, as these eigenvectors u_i satisfy \Sigma u_i = \lambda_i u_i, where \lambda_i are the eigenvalues representing the amount of variance explained by each direction. However, for typical face images (e.g., d \approx 10^4 to $10^5), the eigen decomposition of this full covariance matrix is computationally prohibitive, requiring O(d^3) operations and substantial memory for such high dimensionality. To make the computation feasible, especially when N \ll d, the eigenvectors are derived from the much smaller N \times N matrix A = \Phi^T \Phi, where \Phi is the d \times N data matrix with columns \phi_i. The eigenvalue decomposition of A provides eigenvectors v_i such that A v_i = \mu_i v_i, and the corresponding eigenfaces are obtained as u_i = \Phi v_i, which are proportional to the true eigenvectors of \Sigma (with eigenvalues related by \lambda_i = \mu_i / N). This approach reduces the complexity to O(N^3), which is practical for modest N (e.g., tens to hundreds of training images). The connection to singular value decomposition (SVD) offers an equivalent and often preferred numerical method for extracting these components. Applying SVD to the centered data matrix yields \Phi = U S V^T, where U is a d \times N orthogonal matrix whose columns are the left singular vectors, S is a diagonal matrix of singular values \sigma_i, and V is N \times N orthogonal. The columns of U, scaled by the singular values, correspond to the eigenfaces u_i, and the eigenvalues of \Sigma are given by \lambda_i = \sigma_i^2 / N. This formulation ensures numerical stability, particularly in implementations using libraries like LAPACK, and directly aligns with the PCA basis without explicit covariance formation. Once computed, the eigenfaces are sorted by their associated eigenvalues in descending order, retaining only the top k (where k < N) dominant components that account for the majority of the data variance, typically 90-99% in face datasets. This ranking prioritizes the most informative directions for subsequent dimensionality reduction.

Applications in Recognition

Face Recognition Pipeline

The face recognition pipeline using eigenfaces begins with the projection of a new input face image, denoted as \Gamma, onto the precomputed face space. This involves subtracting the mean face \Psi from \Gamma and computing the projection coefficients, or weights, \omega_j = u_j^T (\Gamma - \Psi) for each of the k selected eigenfaces u_j, where j = 1, 2, \dots, k and k is typically much smaller than the original image dimensionality (e.g., k = 7 out of hundreds of principal components). These weights form the vector \Omega = [\omega_1, \omega_2, \dots, \omega_k], which represents the input image as coordinates in the low-dimensional face space spanned by the eigenfaces. In the classification stage, the projected vector \Omega is compared to the stored weight vectors \Omega_{\text{training}} from the training set of known faces using the Euclidean distance metric ||\Omega - \Omega_{\text{training}}||. The system identifies the input as the known face whose training vector yields the minimum distance, effectively finding the nearest neighbor in the face space. To distinguish faces from non-faces, a distance threshold \Theta_d is applied: if the minimum distance exceeds \Theta_d, the input is rejected as not belonging to any known class or as a non-face. The pipeline operates in two primary modes: verification and identification. In verification mode, the system performs a one-to-one comparison between the input \Omega and a specific target's \Omega_{\text{target}}, accepting the match if ||\Omega - \Omega_{\text{target}}|| < \Theta_e, where \Theta_e is a class-specific error threshold tuned to balance false positives and negatives. In identification mode, it conducts a one-to-many search across all stored training vectors to find the overall closest match, again using the Euclidean distance and thresholds to confirm or reject the result. Error handling relies on these adjustable thresholds, such as \Theta_d for overall face detection (measuring deviation from the face space, e.g., E = ||\Omega|| if projected onto the origin-shifted space) and \Theta_e for classification, which can achieve high accuracy (e.g., 96%) while allowing rejection of unknowns at rates up to 20%.

Practical Implementation Examples

Practical implementations of eigenfaces typically follow a structured pseudocode outline for training and testing phases, emphasizing efficient matrix operations to handle image data. In the training phase, images are loaded and vectorized into a matrix where each column represents a centered face image; the mean face is subtracted from each, forming centered vectors a_i = x_i - \psi; the covariance matrix is approximated via L = A^T A where A is the matrix of centered vectors (to avoid computing the full d \times d matrix when d > N); of L are computed; and the top k eigenvectors are mapped back to the original space as u_l = \sum_{i=1}^N v_l^i a_i to obtain the eigenfaces. For the testing phase, an input image is vectorized and centered (\phi = y - \psi); it is projected onto the eigenfaces to get weights \Omega = [\omega_1, \omega_2, \dots, \omega_k]^T where \omega_j = u_j^T \phi; and occurs by finding the minimum to stored training weights, \epsilon = \min_l \|\Omega - \Omega^l\|, with a to determine matches. Toolkits facilitate these steps through built-in functions for matrix operations and decomposition. In MATLAB, the process leverages eig or pca on the centered image matrix for eigenvector extraction, with image loading via imread and vectorization using reshape. OpenCV supports eigenfaces via cv::PCA::compute on a matrix of flattened grayscale images (e.g., 100x100 pixels), handling I/O with cv::imread and enabling real-time extensions through its face module, though modern versions favor alternatives like LBPH for production. For Python prototyping, scikit-learn's PCA module simplifies implementation by fitting on a 2D array of reshaped images (e.g., pca = PCA(n_components=150); eigenfaces = pca.fit_transform(flattened_faces)), often combined with OpenCV for preprocessing. Integration with public datasets requires handling file I/O and normalization for reproducibility. The Yale Face Database, containing 165 images of 15 subjects under varying lighting and expressions (11 images per subject, cropped to 320x243 pixels), serves as a standard benchmark; images are loaded sequentially, converted to vectors (e.g., via flattening rows), and normalized to zero mean and unit variance before feeding into the training matrix. Computational considerations highlight scalability limits, with time complexity dominated by centering (O(N d)) and covariance approximation (O(N^2 d)), plus eigendecomposition (O(N^3)) for N training images and d pixels per image (typically d ≈ 10^4 for 100x100 images), necessitating GPU acceleration or for datasets exceeding thousands of images on standard hardware. A simple example involves training on 10 images (two per subject for five individuals, vectorized from 64x64 faces), yielding approximately 5 dominant eigenfaces after selecting those with eigenvalues above a variance (e.g., retaining 95% explained variance), enabling basic nearest-neighbor with rates around 80-90% on held-out images from the same subjects under controlled conditions.

Limitations and Modern Context

Performance Critiques

Eigenfaces, as a holistic representation derived from (PCA), exhibits significant sensitivity to environmental and positional variations in face images. The method's reliance on global intensities makes it particularly vulnerable to changes in conditions, where even moderate shifts in illumination can drastically alter the projection coefficients used for . For instance, experiments on datasets with controlled variations demonstrated high , but introducing real-world lighting differences led to error rates as high as 47.7% in extrapolation scenarios, where the training set did not encompass the full range of illumination present in test images. Similarly, occlusions—such as those caused by accessories or partial coverings—and extreme head poses disrupt the approximation, causing the eigenface basis to misalign with the input, as the global features fail to localize discriminative elements like eyes or mouth under such distortions. A key risk in eigenfaces arises from , especially when on limited sample sizes per subject, which is common in early setups. With few images per individual, the principal components tend to capture idiosyncratic or subject-specific artifacts rather than robust, generalizable identity cues, leading to inflated performance on but poor . This small-sample-size problem exacerbates the method's instability, as the becomes dominated by intra-subject variations rather than inter-subject differences, resulting in eigenfaces that are overly tuned to the exact exemplars. Studies have shown that accuracy declines significantly when sample sizes are insufficient, highlighting the need for larger, more diverse datasets to approximate a stable . Empirical evaluations underscore these limitations, contrasting the method's initial promise with its real-world shortcomings. In the original controlled tests using a database of 16 subjects under varied but constrained conditions, eigenfaces achieved approximately 96% accuracy for variations. However, on more challenging benchmarks like the Yale Face Database, which incorporates real-world , expressions, and accessories, the error rate rose to 24.4% with 30 principal components, dropping further to 15.3% only after excluding the first three components that primarily encode illumination. Comparable results on subsets of the FERET database, involving pose and variations, yielded accuracies in the 70-85% range for standard eigenfaces implementations, far below modern standards and underscoring the method's degradation outside ideal settings. These findings from seminal evaluations establish that while eigenfaces performs adequately in uniform environments, its accuracy declines sharply to 70-80% or lower amid practical variations. The linear nature of PCA in eigenfaces also encounters the curse of dimensionality, as human faces inhabit a high-dimensional, nonlinear manifold that linear projections cannot fully capture. This mismatch means that while PCA reduces dimensionality effectively for Gaussian-like distributions, the intrinsic nonlinear structure of face variations—such as nonlinear deformations from poses or expressions—leads to suboptimal representations, increasing misclassification risks in complex scenarios. and concerns further compound these issues, as the method's simplicity, based solely on 2D intensity patterns without liveness verification, renders it highly susceptible to spoofing attacks using printed photos, videos, or masks, where basic implementations lack anti-spoofing measures.

Enhancements and Contemporary Alternatives

To address the limitations of eigenfaces in handling variations such as lighting changes, hybrid approaches have integrated local feature descriptors. One notable enhancement combines eigenfaces with the Weber Local Descriptor (WLD), which preprocesses images to normalize illumination by capturing local texture patterns through differential excitation and gradient orientation. This method applies WLD before (KPCA) and (LDA), followed by a Ridge Classifier for robust matching. On the Extended Yale B dataset, which tests lighting variance, this hybrid achieved 99.83% accuracy, a substantial improvement over the original eigenfaces' 5.63%. Kernel PCA variants extend eigenfaces beyond linear subspaces to handle non-linear face manifolds more effectively. Kernel eigenfaces map face images into a high-dimensional feature space using non-linear kernels, such as kernels of degree 2 or 3, to capture higher-order correlations without explicit computation of the mapping. This non-linear projection outperforms linear by modeling complex facial variations, reducing error rates on benchmarks like the Yale face database from 28.49% (eigenfaces) to 24.24% with a cubic kernel, and on the database to 2.00% with a kernel. Contemporary face recognition has largely shifted from eigenfaces to methods, which offer superior accuracy and robustness. Convolutional Neural Networks (CNNs), exemplified by FaceNet introduced in 2015, embed faces into a using for verification, achieving 99.63% accuracy on the Labeled Faces in the Wild (LFW) dataset—a for unrestricted conditions. More recent advancements incorporate transformers, such as Vision Transformers (ViT) and Swin Transformers, which process global dependencies in facial features for enhanced performance; a 2024 comparative study found transformer-based models outperforming CNNs in tasks like cross-pose recognition, with accuracies exceeding 98% on datasets like FER-2013. These transformer architectures, often hybridized with CNNs, enable real-time recognition in 2024-2025 applications by leveraging attention mechanisms for efficient feature extraction. As of 2025, eigenfaces serve primarily as an educational tool and foundational technique in resource-constrained systems, where their computational simplicity remains advantageous, but they have been supplanted in commercial by due to higher accuracy demands. A 2024 IEEE review highlights ongoing enhancements, yet underscores the method's evolution toward integration with modern pipelines rather than standalone use. Similarly, 2025 on transformer-based emphasizes eigenfaces' historical role while noting their obsolescence in high-stakes applications like .

References

  1. [1]
    Eigenfaces for Recognition | Journal of Cognitive Neuroscience
    Jan 1, 1991 · We have developed a near-real-time computer system that can locate and track a subject's head, and then recognize the person by comparing characteristics of ...
  2. [2]
    Eigenfaces - Scholarpedia
    Sep 24, 2012 · The Eigenface approach is considered by many to be the first working facial recognition technology, and it served as the basis for one of the ...
  3. [3]
    Eigenfaces for Recognition - MIT Media Lab
    Dec 1, 1991 · Eigenfaces for ... face images. The significant features are known as "eigenfaces," because they are the eigenvectors (principal components) ...
  4. [4]
    Principal Components Analysis
    The central idea of principal component analysis (PCA) is to reduce the dimensionality of a data set consisting of a large number of interrelated variables.
  5. [5]
    Eigenface Approach - an overview | ScienceDirect Topics
    ... pose and lighting changes. However, their ... Because they appear as somewhat ghostly faces, as shown in Figure 2.3 , they are called eigenfaces.
  6. [6]
    A Brief History of Face Recognition - FaceFirst
    Aug 1, 2017 · As with Bledsoe's system, the actual biometrics had to still be manually computed. Eigenfaces (Late 1980s-Early 1990s). In 1988, ...
  7. [7]
    Google Scholar
    No readable text found in the HTML.<|control11|><|separator|>
  8. [8]
    Past, Present, and Future of Face Recognition: A Review - MDPI
    Mathematically, the eigenfaces represent the facial distribution's main components or the eigenvectors of the facial image set covariance matrix.<|control11|><|separator|>
  9. [9]
    The FERET database and evaluation procedure for face-recognition ...
    The FERET database is a large facial image database, and its evaluation procedure is an independent test to compare and assess face-recognition algorithms.
  10. [10]
    face recognition using pca and eigen face approach - ResearchGate
    May 4, 2017 · ... 92x112, 8-bit grey levels. The images are organised. in 40 ... eigenfaces. Figure 6: Last three Eigen face. Figure 7.– Euclidean ...
  11. [11]
    [PDF] FACE RECOGNITION USING PCA AND EIGEN FACE APPROACH
    The size of each image is 92x112, 8-bit grey levels. The images are ... All im- ages are grayscale (intensity levels of gray are taken as image ...
  12. [12]
    (PDF) Facial recognition using eigenfaces by PCA - ResearchGate
    The present paper refers to different face recognition approaches and primarily focuses on principal component analysis.Missing: origins | Show results with:origins
  13. [13]
    [PDF] Evaluation of Face R Recognition Techniques for Application to ...
    This paper evaluates face recognit real-world application of Facebook usually present results in terms constrained face datasets, it is difficult.Missing: influence | Show results with:influence
  14. [14]
    Facial Recognition through Eigenfaces - twopointseven
    Oct 5, 2017 · The first step when dealing with face images is to represent each image Ik as a column vector of grayscale values for each pixel. This process ...<|control11|><|separator|>
  15. [15]
    Eigenface - an overview | ScienceDirect Topics
    Eigenfaces is defined as a method for face recognition that utilizes Principal Component Analysis (PCA) to extract characteristic features of faces, ...
  16. [16]
    Eigenfaces reconstructions - Bastian Grossenbacher Rieck
    Feb 28, 2024 · Treat each digital image as a vector in a high-dimensional space by “flattening” the image vector. ... Turk and Pentland call the “face space”.
  17. [17]
    [PDF] Eigenfaces for Recognition - Curses at DTU Compute
    We have developed a near-real-time computer system that can locate and track a subject's head, and then recognize the person by comparing characteristics of ...
  18. [18]
    [PDF] Recognition, SVD, and PCA - cs.Princeton
    • Compute pseudoinverse using SVD. – Lets you see if data is singular. – Even ... PCA on Faces: “Eigenfaces”. Average face. First principal component. Other.
  19. [19]
    ML | Face Recognition Using Eigenfaces (PCA Algorithm)
    Sep 24, 2021 · In 1991, Turk and Pentland suggested an approach to face recognition that uses dimensionality reduction and linear algebra concepts to ...
  20. [20]
    [PDF] Eigenfaces for Recognition: Matthew Turk and Alex Pentland
    The idea of using eigenfaces was motivated by a tech- nique developed by Sirovich and Kirby (1987) and Kirby and Sirovich (1990) for efficiently representing ...
  21. [21]
    eigenfaces algorithm - File Exchange - MATLAB Central - MathWorks
    Mar 17, 2014 · 1) resize all M faces to N*N · 2) remove average · 3) create matrix A of faces each row N*N · 4) calculate average face · 5) remove average face ...Missing: example | Show results with:example
  22. [22]
    Face Recognition with OpenCV
    The Eigenfaces method described in [218] took a holistic approach to face recognition: A facial image is a point from a high-dimensional image space and a lower ...
  23. [23]
  24. [24]
    [PDF] 2D Face Recognition Using PCA, ICA and LDA Shireen Elhabian ...
    Database of Faces. The Yale Face Database [1] consists of 165 grayscale images from 15 individuals. There are 11 images per person, with one image per face ...
  25. [25]
    [PDF] Pattern Recognition with Eigenfaces
    The total time complexity of low-dimensional PCA on a. D× N dataset (including matrix multiplications, eigenvalue decomposition and other steps) is O(Nq1D) 2.
  26. [26]
    [PDF] Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear ...
    The Eigenface technique, another method based on linearly projecting the image space to a low dimensional subspace, has similar computational requirements.
  27. [27]
    Linear discriminant analysis for the small sample size problem
    Aug 10, 2025 · The limited sample size likely led to overfitting, where the ML model memorizes patterns within the dataset too closely, and therefore fails to ...
  28. [28]
    None
    ### Summary of Face Recognition Pipeline Using Eigenfaces (Turk & Pentland, MIT)
  29. [29]
    Face recognition using view-based and modular eigenspaces
    ... eigenfaces for recognition and interactive search in the FERET face database. A recognition accuracy ... This figure is consistent with the 95% recognition rate ...
  30. [30]
  31. [31]
    Face recognition using kernel eigenfaces - IEEE Xplore
    Eigenface or principal component analysis (PCA) methods have demonstrated their success in face recognition, detection, and tracking.
  32. [32]
    FaceNet: A Unified Embedding for Face Recognition and Clustering
    Mar 12, 2015 · On the widely used Labeled Faces in the Wild (LFW) dataset, our system achieves a new record accuracy of 99.63%. On YouTube Faces DB it ...
  33. [33]
    Comprehensive comparison between vision transformers and ...
    This paper presents a comprehensive comparison between Vision Transformers and Convolutional Neural Networks for face recognition related tasks.
  34. [34]
    Face Detection Using Eigenfaces: A Comprehensive Review
    Jul 30, 2024 · This paper reviews face detection using Eigenfaces, a method rooted in PCA, exploring its advancements, challenges, and prospects.
  35. [35]
    Analyzing the Current Status of the Transformer Model for a Face ...
    This paper systematically analyzes the evolution path of face recognition technology based on the Transformer architecture, focusing on the four technical ...