Self-similarity representation of Weber faces for kinship classification
06 Dec 2012-pp 245-250
TL;DR: A kinship classification algorithm that uses the local description of the pre-processed Weber face image to outperforms an existing algorithm and yields a classification accuracy of 75.2%.
Abstract: Establishing kinship using images can be utilized as context information in different applications including face recognition. However, the process of automatically detecting kinship in facial images is a challenging and relatively less explored task. The reason for this includes limited availability of datasets as well as the inherent variations amongst kins. This paper presents a kinship classification algorithm that uses the local description of the pre-processed Weber face image. A kinship database is also prepared that contains images pertaining to 272 kin pairs. The database includes images of celebrities (and their kins) and has four ethnicity groups and seven kinship groups. The proposed algorithm outperforms an existing algorithm and yields a classification accuracy of 75.2%.
Citations
More filters
01 Jan 2014
TL;DR: This thesis focuses on temporal changes in the face, and dynamic patterns of expressions, which may lead to more efficient uses of the temporal information and dynamic features in face processing and affective computing.
Abstract: Most of the approaches in automatic face analysis rely solely on static appearance. However, temporal analysis of expressions reveals interesting patterns. For a better understanding of the human face, this thesis focuses on temporal changes in the face, and dynamic patterns of expressions. In addition to improving the state of the art in several areas of automatic face analysis, the present thesis introduces new and significant findings on facial dynamics. The contributions on temporal analysis and understanding of faces can be summarized as follows: 1) An accurate facial landmarking method is proposed to enable detailed analysis of facial movements; 2) Dynamic feature descriptors are introduced to reveal the temporal patterns of facial expressions; 3) Various frameworks are proposed to exploit temporal information and facial dynamics in expression spontaneity analysis, age estimation, and kinship verification; 4) An affect-responsive system is designed to create an adaptive application empowered by face-to-face human-computer interaction. We believe that affective technologies will shape the future by providing a more natural form of human-machine interaction. To this end, the proposed methods and ideas may lead to more efficient uses of the temporal information and dynamic features in face processing and affective computing.
10 citations
01 Mar 2014
TL;DR: A survey on how to verify family relation by a various metric learning method, probabilistic framework and several methods which can extract critical points on a face using both location and texture information such as lip corners, eye corners and nose tip.
Abstract: Verification of Family Relationship from facial images is a challenging problem in computer vision, and there are very few attempts on tackling this problem in the literature. We present a survey on how to verify family relation by a various metric learning method, probabilistic framework and several methods which can extract critical points on a face using both location and texture information such as lip corners, eye corners and nose tip. These are critical points in a human face, and centre of the mouth of each face. Motivated by the fact that interclass samples (without kinship relations) with higher similarity usually lie in a neighborhood and are more easily misclassified than those with lower similarity, we aim to learn a distance metric under which the interclass samples (with kinship relations) are pushed as close as possible and interclass samples lying in a neighborhood are repulsed and pulled as far as possible, simultaneously, such that more discriminative information can be exploited for verification. To make better use of multiple feature descriptors to further improve the verification performance it is necessary to develop a method to extract the salient familial traits in face images for kinship recognition. If this idea works, an instrument may be invented to measure familial relationships. This computational kinship measurement might have a large impact in real applications, such as child adoptions, trafficking/smuggling of children, and finding missing children, identifying relatives from a photo collection. We can collect Dataset of young parent and old parent face images from Internet. In our research work verification has to be performed by measuring the number of image pairs available for training and testing along with images of their parents and children frontal images. These images are taken as kinship so as to maintain relationship between two persons who are biologically related with overlapping genes.
9 citations
18 May 2015
TL;DR: This work tackles the problem of tri-subjects kinship verification by effectively exploiting the dependence structure between child and parents in a few aspects: similarity measure, feature selection and classifier design.
Abstract: Recent research has demonstrated that computer vision algorithms have understood individual face image fairly well However, one major challenge in computer vision is to go beyond that and to investigate the bi-or tri- relationship among multiple visual entities, answering such questions as whether a child in a photo belongs to given parents Indeed parents-child relationship plays a core role in a family and understanding such kind of kin relationship would have fundamental impact on the behavior of an artificial intelligent agent working in a human being world In this work, we tackle the problem of tri-subjects kinship verification by effectively exploiting the dependence structure between child and parents in a few aspects: similarity measure, feature selection and classifier design State-of-the-art results are reported with the proposed method on our newly released kinship database characterized by over 1,000 parents-child groups
9 citations
Additional excerpts
...Other feature sets for kinship verification include Gradient Orientation Pyramid (GGOP) [8], Self Similarity Representation (SSR) [25] and prototype-based discriminative feature learning (PDFL) method [26]....
[...]
...tation (SSR) [26] and prototype-based discriminative feature...
[...]
TL;DR: A comprehensive review of the state-of-the-art methods for Facial Kinship Verification (FKV) can be found in this paper , where the authors identify gaps in current research and discuss potential future research directions.
Abstract: Abstract The goal of Facial Kinship Verification (FKV) is to automatically determine whether two individuals have a kin relationship or not from their given facial images or videos. It is an emerging and challenging problem that has attracted increasing attention due to its practical applications. Over the past decade, significant progress has been achieved in this new field. Handcrafted features and deep learning techniques have been widely studied in FKV. The goal of this paper is to conduct a comprehensive review of the problem of FKV. We cover different aspects of the research, including problem definition, challenges, applications, benchmark datasets, a taxonomy of existing methods, and state-of-the-art performance. In retrospect of what has been achieved so far, we identify gaps in current research and discuss potential future research directions.
8 citations
11 Dec 2014
TL;DR: Experimental results have shown that the proposed system can effectively annotate the verification of family relation and observe in the experiments that LBP features perform stably and robustly over a useful range of less resolutions of facial images.
Abstract: There are many social networking web sites used by people and number of photos is uploaded by them. But from photos it is difficult to predict the relationship among the people if necessary. So there is need of system for automatic identification and prediction of relationship among them, specifically kinship from photo. So, we proposed system, which uses Computer Vision, Face recognition, Feature extraction and classification to solve this problem. Proposed System first detects all the features from given photo then extracts them from the faces using Local binary Pattern. Our analysis and psychological studies show that the facial resemblance differs from member to member and depends on image segmentation and image histogram. Implementing the proposed approach on the collected limited family database from Kinface V2 and Family 101 dataset contain child, father and mother images. Our dataset contain family images of Indian celebrities achieved considerable improvement. This computational kinship measurement have a large impact in real applications such as child adoptions, trafficking/smuggling of children, and finding missing children, identifying relatives from a photo collection. So, we empirically evaluate facial representation based on statistical local features, Local Binary Pattern. One can also find classification of the feature, KNN with Euclidean distance find minimum distance from histogram sequence. We observe in our experiments that LBP features perform stably and robustly over a useful range of less resolutions of facial images. We proposed an algorithm to predict the most likely kin relationships embedded in an image from three input images of child, mother and father. In addition, human subjects are used in a baseline study on three databases. Experimental results have shown that the proposed system can effectively annotate the verification of family relation. One way for family (kinship) verification is to perform DNA test that is currently accurate. The DNA test is not suitable for mass screening and a very expensive test for crime scene investigations that takes days to get the results.
8 citations
References
More filters
20 Jun 2005
TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.
Abstract: We study the question of feature sets for robust visual object recognition; adopting linear SVM based human detection as a test case. After reviewing existing edge and gradient based descriptors, we show experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection. We study the influence of each stage of the computation on performance, concluding that fine-scale gradients, fine orientation binning, relatively coarse spatial binning, and high-quality local contrast normalization in overlapping descriptor blocks are all important for good results. The new approach gives near-perfect separation on the original MIT pedestrian database, so we introduce a more challenging dataset containing over 1800 annotated human images with a large range of pose variations and backgrounds.
31,952 citations
"Self-similarity representation of W..." refers background in this paper
...75% which was comparatively better than Local Binary Pattern (LBP) [1], Histogram of Gradients (HOG) [4], Principal Component Analysis (PCA) [8], and Linear Embedding (LE) [2]....
[...]
...The authors reported an accuracy of 67.75% which was comparatively better than Local Binary Pattern (LBP) [1], Histogram of Gradients (HOG) [4], Principal Component Analysis (PCA) [8], and Linear Embedding (LE) [2]....
[...]
01 Jan 1998
TL;DR: Presenting a method for determining the necessary and sufficient conditions for consistency of learning process, the author covers function estimates from small data pools, applying these estimations to real-life problems, and much more.
Abstract: A comprehensive look at learning and generalization theory. The statistical theory of learning and generalization concerns the problem of choosing desired functions on the basis of empirical data. Highly applicable to a variety of computer science and robotics fields, this book offers lucid coverage of the theory as a whole. Presenting a method for determining the necessary and sufficient conditions for consistency of learning process, the author covers function estimates from small data pools, applying these estimations to real-life problems, and much more.
26,531 citations
"Self-similarity representation of W..." refers methods in this paper
...The χ(2) distance measures (in a vector form) are provided as input to the SVM classifier [11] with the classes being kin and non-kin....
[...]
01 Dec 2001
TL;DR: A machine learning approach for visual object detection which is capable of processing images extremely rapidly and achieving high detection rates and the introduction of a new image representation called the "integral image" which allows the features used by the detector to be computed very quickly.
Abstract: This paper describes a machine learning approach for visual object detection which is capable of processing images extremely rapidly and achieving high detection rates. This work is distinguished by three key contributions. The first is the introduction of a new image representation called the "integral image" which allows the features used by our detector to be computed very quickly. The second is a learning algorithm, based on AdaBoost, which selects a small number of critical visual features from a larger set and yields extremely efficient classifiers. The third contribution is a method for combining increasingly more complex classifiers in a "cascade" which allows background regions of the image to be quickly discarded while spending more computation on promising object-like regions. The cascade can be viewed as an object specific focus-of-attention mechanism which unlike previous approaches provides statistical guarantees that discarded regions are unlikely to contain the object of interest. In the domain of face detection the system yields detection rates comparable to the best previous systems. Used in real-time applications, the detector runs at 15 frames per second without resorting to image differencing or skin color detection.
18,620 citations
"Self-similarity representation of W..." refers methods in this paper
...The face region present in the image is first extracted using the Adaboost face detector [12]....
[...]
Book•
03 Oct 1988TL;DR: This chapter discusses two Dimensional Systems and Mathematical Preliminaries and their applications in Image Analysis and Computer Vision, as well as image reconstruction from Projections and image enhancement.
Abstract: Introduction. 1. Two Dimensional Systems and Mathematical Preliminaries. 2. Image Perception. 3. Image Sampling and Quantization. 4. Image Transforms. 5. Image Representation by Stochastic Models. 6. Image Enhancement. 7. Image Filtering and Restoration. 8. Image Analysis and Computer Vision. 9. Image Reconstruction From Projections. 10. Image Data Compression.
8,504 citations
"Self-similarity representation of W..." refers methods in this paper
...Therefore, Difference of Gaussian (DoG) approach [7] has been applied to extract these features using the steps below....
[...]
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
"Self-similarity representation of W..." refers background in this paper
...75% which was comparatively better than Local Binary Pattern (LBP) [1], Histogram of Gradients (HOG) [4], Principal Component Analysis (PCA) [8], and Linear Embedding (LE) [2]....
[...]
...The authors reported an accuracy of 67.75% which was comparatively better than Local Binary Pattern (LBP) [1], Histogram of Gradients (HOG) [4], Principal Component Analysis (PCA) [8], and Linear Embedding (LE) [2]....
[...]