The paper “Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification” by Joy Buolamwini and Timnit Gebru, that will be presented at the Conference on Fairness, Accountability, and Transparency (FAT*) in February 2018, evaluates three commercial API-based classifiers of gender from facial images, including IBM Watson Visual Recognition. The study finds these services to have recognition capabilities that are not balanced over genders and skin tones [1]. In particular, the authors show that the highest error involves images of dark-skinned women, while the most accurate result is for light-skinned men.

/pdf/gender-shades-intersectional-accuracy-disparities-in-4qgeu0c1i3.pdf

Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification

Pattern recognition

In recent decades, we have witnessed the evolution of biometric technology from the first pioneering works in face and voice recognition to the current state of development wherein a wide spectrum of highly accurate systems may be found, ranging from largely deployed modalities, such as fingerprint, face, or iris, to more marginal ones, such as signature or hand. This path of technological evolution has naturally led to a critical issue that has only started to be addressed recently: the resistance of this rapidly emerging technology to external attacks and, in particular, to spoofing. Spoofing, referred to by the term presentation attack in current standards, is a purely biometric vulnerability that is not shared with other IT security solutions. It refers to the ability to fool a biometric system into recognizing an illegitimate user as a genuine one by means of presenting a synthetic forged version of the original biometric trait to the sensor. The entire biometric community, including researchers, developers, standardizing bodies, and vendors, has thrown itself into the challenging task of proposing and developing efficient protection methods against this threat. The goal of this paper is to provide a comprehensive overview on the work that has been carried out over the last decade in the emerging field of antispoofing, with special attention to the mature and largely deployed face modality. The work covers theories, methodologies, state-of-the-art techniques, and evaluation databases and also aims at providing an outlook into the future of this very active field of research.

Biometric Antispoofing Methods: A Survey in Face Recognition

Face recognition systems are gaining momentum with current developments in computer vision. At the same time, tactics to mislead these systems are getting more complex, and counter-measure approaches are necessary. Following the current progress with convolutional neural networks (CNN) in classification tasks, we present an approach based on transfer learning using a pre-trained CNN model using only static features to recognize photo, video or mask attacks. We tested our approach on the REPLAY-ATTACK and 3DMAD public databases. On the REPLAY-ATTACK database our accuracy was 99.04% and the half total error rate (HTER) of 1.20%. For the 3DMAD, our accuracy was of 100.00% and HTER 0.00%. Our results are comparable to the state-of-the-art.

Transfer Learning Using Convolutional Neural Networks for Face Anti-spoofing

An image is worth a thousand words; hence, a face image illustrates extensive details about the specification, gender, age, and emotional states of mind. Facial expressions play an important role in community-based interactions and are often used in the behavioral analysis of emotions. Recognition of automatic facial expressions from a facial image is a challenging task in the computer vision community and admits a large set of applications, such as driver safety, human–computer interactions, health care, behavioral science, video conferencing, cognitive science, and others. In this work, a deep-learning-based scheme is proposed for identifying the facial expression of a person. The proposed method consists of two parts. The former one finds out local features from face images using a local gravitational force descriptor, while, in the latter part, the descriptor is fed into a novel deep convolution neural network (DCNN) model. The proposed DCNN has two branches. The first branch explores geometric features, such as edges, curves, and lines, whereas holistic features are extracted by the second branch. Finally, the score-level fusion technique is adopted to compute the final classification score. The proposed method along with 25 state-of-the-art methods is implemented on five benchmark available databases, namely, Facial Expression Recognition 2013, Japanese Female Facial Expressions, Extended CohnKanade, Karolinska Directed Emotional Faces, and Real-world Affective Faces. The databases consist of seven basic emotions: neutral, happiness, anger, sadness, fear, disgust, and surprise. The proposed method is compared with existing approaches using four evaluation metrics, namely, accuracy, precision, recall, and f1-score. The obtained results demonstrate that the proposed method outperforms all state-of-the-art methods on all the databases.

Facial Expression Recognition Using Local Gravitational Force Descriptor-Based Deep Convolution Neural Networks

Thermal infrared (IR) images focus on changes of temperature distribution on facialmuscles and blood vessels. These temperature changes can be regarded as texture features of images. A comparative study of face two recognition methods working in thermal spectrum is carried out in this paper. In the first approach, the training images and the test images are processed with Haar wavelet transform and the LL band and the average of LH/HL/HH bands subimages are created for each face image. Then a total confidence matrix is formed for each face image by taking a weighted sum of the corresponding pixel values of the LL band and average band. For LBP feature extraction, each of the face images in training and test datasets is divided into 161 numbers of subimages, each of size 8 × 8 pixels. For each such subimages, LBP features are extracted which are concatenated in manner. PCA is performed separately on the individual feature set for dimensionality reduction. Finally, two different classifiers namely multilayer feed forward neural network and minimum distance classifier are used to classify face images. The experiments have been performed on the database created at our own laboratory and Terravic Facial IR Database.

/pdf/comparative-study-of-human-thermal-face-recognition-based-on-23p0z9pio4.pdf

Comparative study of human thermal face recognition based on Haar wavelet transform and local binary pattern

In this paper, an efficient approach for human face recognition based on the use of minutiae points in thermal face image is proposed. The thermogram of human face is captured by thermal infra-red camera. Image processing methods are used to pre-process the captured thermogram, from which different physiological features based on blood perfusion data are extracted. Blood perfusion data are related to distribution of blood vessels under the face skin. In the present work, three different methods have been used to get the blood perfusion image, namely bit-plane slicing and medial axis transform, morphological erosion and medial axis transform, sobel edge operators. A five layer feed-forward back propagation neural network is used as the classification tool. It has been found that the first method supercedes the other two producing an accuracy of 97.62% with block size 16×16 for bit-plane 4.

Automated thermal face recognition based on minutiae extraction

In this paper, we present a novel approach for three-dimensional face recognition by extracting the curvature maps from range images. There are four types of curvature maps: Gaussian, Mean, Maximum and Minimum curvature maps. These curvature maps are used as a feature for 3D face recognition purpose. The dimension of these feature vectors is reduced using Singular Value Decomposition (SVD) technique. Now from calculated three components of SVD, the nonnegative values of ‘S’ part of SVD is ranked and used as feature vector. In this proposed method, two pair-wise curvature computations are done. One is Mean, and Maximum curvature pair and another is Gaussian and Mean curvature pair. These are used to compare the result for better recognition rate. This automated 3D face recognition system is focused in different directions like, frontal pose with expression and illumination variation, frontal face along with registered face, only registered face and registered face from different pose orientation across X, Y and Z axes. 3D face images used for this research work are taken from FRAV3D database. The pose variation of 3D facial image is being registered to frontal pose by applying one to all registration technique then curvature mapping is applied on registered face images along with remaining frontal face images. For the classification and recognition purpose five layer feed-forward back propagation neural network classifiers is used, and the corresponding result is discussed in section 4.

3d face recognition from range images based on curvature analysis

This paper presents an algorithm for feature selection based on particle swarm optimization (PSO) for thermal face recognition. The total algorithm goes through many steps. In the very first step, thermal human face image is preprocessed and cropping of the facial region from the entire image is done. In the next step, scale invariant feature transform (SIFT) is used to extract the features from the cropped face region. The features obtained by SIFT are invariant to object rotation and scale. But some irrelevant and noisy features could be produced with the actual features. Unwanted features have to be removed. In other words, optimum features have to be selected for better recognition accuracy. The PSO helps to identify the optimum features set using local as well as global searches. Here, this process has been implemented to select a subset of features that effectively represents original feature extracted for better classification convergence. Finally, minimum distance classifier is used to find the class label of each testing images. Minimum distance classifier acts as an objective function for PSO. In this work, all the experiments have been performed on UGC-JU thermal face database. The maximum success rate of 98.61 % recognition has been achieved using SIFT and PSO for frontal face images and 90.28 % for all images.

https://www.springer.com/cda/content/document/cda_downloaddocument/9788132219842-c2.pdf?SGWID=0-0-45-1474353-p176821617

Feature Selection using Particle Swarm Optimization for Thermal Face Recognition

Thermal infra-red (IR) images focus on changes of temperature distribution on facial muscles and blood vessels. These temperature changes can be regarded as texture features of images. A comparative study of face recognition methods working in thermal spectrum is carried out in this paper. In these study two local-matching methods based on Haar wavelet transform and Local Binary Pattern (LBP) are analyzed. Wavelet transform is a good tool to analyze multi-scale, multi-direction changes of texture. Local binary patterns (LBP) are a type of feature used for classification in computer vision. Firstly, human thermal IR face image is preprocessed and cropped the face region only from the entire image. Secondly, two different approaches are used to extract the features from the cropped face region. In the first approach, the training images and the test images are processed with Haar wavelet transform and the LL band and the average of LH/HL/HH bands sub-images are created for each face image. Then a total confidence matrix is formed for each face image by taking a weighted sum of the corresponding pixel values of the LL band and average band. For LBP feature extraction, each of the face images in training and test datasets is divided into 161 numbers of sub images, each of size 8X8 pixels. For each such sub images, LBP features are extracted which are concatenated in row wise manner. PCA is performed separately on the individual feature set for dimensionality reeducation. Finally two different classifiers are used to classify face images. One such classifier multi-layer feed forward neural network and another classifier is minimum distance classifier. The Experiments have been performed on the database created at our own laboratory and Terravic Facial IR Database.

Suranjan Ganguly

Papers

Comparative study of human thermal face recognition based on Haar wavelet transform and local binary pattern

Automated thermal face recognition based on minutiae extraction

3d face recognition from range images based on curvature analysis

Feature Selection using Particle Swarm Optimization for Thermal Face Recognition

A Comparative Study of Human thermal face recognition based on Haar wavelet transform (HWT) and Local Binary Pattern (LBP)