Face recognition for look-alikes: A preliminary study
TL;DR: The analysis shows that neither humans nor automatic face recognition algorithms are efficient in recognizing look-alikes, and an algorithm is proposed to improve the face verification accuracy.
Abstract: One of the major challenges of face recognition is to design a feature extractor and matcher that reduces the intraclass variations and increases the inter-class variations. The feature extraction algorithm has to be robust enough to extract similar features for a particular subject despite variations in quality, pose, illumination, expression, aging, and disguise. The problem is exacerbated when there are two individuals with lower inter-class variations, i.e., look-alikes. In such cases, the intra-class similarity is higher than the inter-class variation for these two individuals. This research explores the problem of look-alike faces and their effect on human performance and automatic face recognition algorithms. There is three fold contribution in this research: firstly, we analyze the human recognition capabilities for look-alike appearances. Secondly, we compare human recognition performance with ten existing face recognition algorithms, and finally, proposed an algorithm to improve the face verification accuracy. The analysis shows that neither humans nor automatic face recognition algorithms are efficient in recognizing look-alikes.
Citations
More filters
01 Sep 2021
TL;DR: In this article, a baseline measure of facial similarity between identical twins and applying this similarity measure to determine the impact of doppelgangers, or look-alikes, on automated facial recognition performance for large face datasets is presented.
Abstract: The problem of distinguishing identical twins and non-twin look-alikes in automated facial recognition (FR) applications has become increasingly important with the widespread adoption of facial biometrics. This work presents an application of one of the largest twin datasets compiled to date to address two FR challenges: 1) determining a baseline measure of facial similarity between identical twins and 2) applying this similarity measure to determine the impact of doppelgangers, or look-alikes, on FR performance for large face datasets. The facial similarity measure is determined via a deep convolutional neural network. The proposed network provides a quantitative similarity score for any two given faces and has been applied to large-scale face datasets to identify similar face pairs.
7 citations
TL;DR: This paper proposes a cohort selection method called K-medoids Cohort Selection (KMCS) to select a reference set of non-matched templates which are almost appropriate to the respective subjects.
Abstract: Face recognition is itself a very challenging task and it becomes more challenging when the input images have intra class variations and inter class similarities in a large scale. Yet the recognition accuracy can be improved in some extent by supporting the system with non-matched templates. Therefore a set of cohort images is used in this regard. But all the cohort templates of the initial cohort pool may not be relevant for each and every enrolled subject. So the main focus of this work is to select a subject specific and meaningful cohort subset. This paper proposes a cohort selection method called K-medoids Cohort Selection (KMCS) to select a reference set of non-matched templates which are almost appropriate to the respective subjects. Basically, all cohort scores of a subject are clustered first using K-medoids clustering. Afterward the cluster having more scattered members/scores from its medoid is selected as a cohort subset because this cluster is constituted with the cohorts carrying more discriminative features compared to others. The SIFT points and SURF points are extracted as facial feature. The experiments are conducted on FEI, ORL and Look-alike databases of face images. The matching scores between probe and query images are normalized using T-norm, Max-Min and Aggarwal (Max rule) cohort score normalization techniques before taking the final decision of acceptance or rejection. The results obtained from the experiments show the domination of the proposed system over the non-cohort face recognition system as well as random and Top 10 cohort selection methods. There is another comparative study between k-means and K-medoids clustering for cohort selection.
6 citations
01 Jan 2013
TL;DR: The concept of biometric mixing has several benefits and can be easily incorporated into existin g biometric systems, and is demonstrated in two different applica tions.
Abstract: Mixing Biometric Data For Generating Joint Identities and P reserving Privacy by Asem A. Othman Doctor of Philosophy in Electrical Engineering West Virginia University Arun A. Ross, Ph.D., Chair Biometrics is the science of automatically recognizing ind ividuals by utilizing biological traits such as fingerprints, face, iris and voice. A classical biome tric system digitizes the human body and uses this digitized identity for human recognition. In t his work, we introduce the concept of mixing biometrics. Mixing biometrics refers to the process of generating a new biometric image by fusing images of different fingers, different faces, or di fferent irises. The resultant mixed image can be used directly in the feature extraction and matc hing stages of an existing biometric system. In this regard, we design and systematically evalua te novel methods for generating mixed images for the fingerprint, iris and face modalities. Furthe r, we extend the concept of mixing to accommodate two distinct modalities of an individual, viz. , fingerprint and iris. The utility of mixing biometrics is demonstrated in two different applica tions. The first application deals with the issue of generating a joint digital identity. A joint ide ntity inherits its uniqueness from two or more individuals and can be used in scenarios such as joint bank accounts or two-man rule systems. The second application deals with the issue of biom etric privacy, where the concept of mixing is used for de-identifying or obscuring biometric im ages and for generating cancelable biometrics. Extensive experimental analysis suggests tha the concept of biometric mixing has several benefits and can be easily incorporated into existin g biometric systems.
5 citations
14 May 2016
TL;DR: The challenges that were encountered while designing the game, steps that were taken to overcome these challenges and results of the preliminary evaluation of the current game design are outlined.
Abstract: This paper describes the on-going work of developing a serious game (a.k.a "game with a purpose") to solve the NP-hard problem of n-way merging. We outline the challenges that were encountered while designing the game, steps that we took to overcome these challenges and results of the preliminary evaluation of our current game design. We hope our experience will be useful for those developing serious games to solve other computationally expensive problems.
5 citations
30 Oct 2020
TL;DR: In this paper, a phased research report on the development of face recognition technology on this basis is presented, where the progress of face classification research in various countries is discussed, and the authors mainly analyze the domestic situation.
Abstract: At present, face recognition technology is widely used in various fields, which effectively promotes the process of information automation and data fusion. Face recognition technology is based on in-depth analysis of image processing technology and high-precision control of face dynamic changes. Identification function. To fully expand the application fields of face recognition technology, according to the humanistic depth and regional environmental characteristics in a certain period, this article will form a phased research report on the development of face recognition technology on this basis. The progress of face classification research in various countries. The report mainly analyzes the domestic situation. Summarized the current status of foreign research, analyzed the differences in facial features in some countries, and analyzed the current common facial feature extraction methods and the advantages and disadvantages of several research methods.
4 citations
References
More filters
TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Abstract: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and are shown to provide robust matching across a substantial range of affine distortion, change in 3D viewpoint, addition of noise, and change in illumination. The features are highly distinctive, in the sense that a single feature can be correctly matched with high probability against a large database of features from many images. This paper also describes an approach to using these features for object recognition. The recognition proceeds by matching individual features to a database of features from known objects using a fast nearest-neighbor algorithm, followed by a Hough transform to identify clusters belonging to a single object, and finally performing verification through least-squares solution for consistent pose parameters. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.
46,906 citations
TL;DR: A novel scale- and rotation-invariant detector and descriptor, coined SURF (Speeded-Up Robust Features), which approximates or even outperforms previously proposed schemes with respect to repeatability, distinctiveness, and robustness, yet can be computed and compared much faster.
Abstract: This article presents a novel scale- and rotation-invariant detector and descriptor, coined SURF (Speeded-Up Robust Features). SURF approximates or even outperforms previously proposed schemes with respect to repeatability, distinctiveness, and robustness, yet can be computed and compared much faster. This is achieved by relying on integral images for image convolutions; by building on the strengths of the leading existing detectors and descriptors (specifically, using a Hessian matrix-based measure for the detector, and a distribution-based descriptor); and by simplifying these methods to the essential. This leads to a combination of novel detection, description, and matching steps. The paper encompasses a detailed description of the detector and descriptor and then explores the effects of the most important parameters. We conclude the article with SURF's application to two challenging, yet converse goals: camera calibration as a special case of image registration, and object recognition. Our experiments underline SURF's usefulness in a broad range of topics in computer vision.
12,449 citations
TL;DR: This paper presents a novel and efficient facial image representation based on local binary pattern (LBP) texture features that is assessed in the face recognition problem under different challenges.
Abstract: This paper presents a novel and efficient facial image representation based on local binary pattern (LBP) texture features. The face image is divided into several regions from which the LBP feature distributions are extracted and concatenated into an enhanced feature vector to be used as a face descriptor. The performance of the proposed method is assessed in the face recognition problem under different challenges. Other applications and several extensions are also discussed
5,563 citations
"Face recognition for look-alikes: A..." refers methods in this paper
...LBP [1] encodes the texture of an image and usesχ2 distance to compute the match scores....
[...]
...The experiments on PCA-KPCA, ICAKICA, LDA-KLDA, LBP, EUCLBP, SURF, and GNN are performed using a large database with different challenging variations on pose, expression and illumination....
[...]
...Among several improvements over LBP, EUCLBP [3] has shown significant improvement....
[...]
...ROC plots for texture descriptor based approaches(LBP, EUCLBP, and SURF) on the look-alike face database. algorithms are not able to discriminate between the inter and intra-class variations....
[...]
...Texture descriptor based algorithms are also used for performance comparison, namely: Local Binary Pattern (LBP), Extended Uniform Circular LBP (EUCLBP), Speeded Up Robust Feature (SURF) descriptors, and dynamic feed-forward neural network architecture based 2D log polar Gabor transform (GNN) [16]....
[...]