Face Image Retrieval Research Articles

Content-based image retrieval based on multiple extended fuzzy-rough framework

This paper presents a content-based image retrieval (CBIR) system with applications in one general purpose and two face image databases using two MPEG-7 image descriptors. The proposed method uses several sophisticated fuzzy-rough feature selection methods and combines the results of these methods to obtain a prominent feature subset for image representation for a particular query. Next, fuzzy-rough upper approximation of the target set (relevant list of images) with respect to the entire database that is represented by the prominent feature subset, is computed for retrieval and ranking. The information table on which every feature selection method works is small in size. Main reasons of performance boost of the proposed method are twofold. One is efficient feature subsets selection. The other reason is the fuzzy-indiscernibility relation based fuzzy-rough framework for computing upper-approximation which supports the approximate equality or similarity sense of CBIR. Fuzzy-rough upper approximation possibly adds more similar images in the relevant list from boundary region to expand the relevant list. The effectiveness of the proposed method is supported by the comparative results obtained from several single dimensionality reduction method, several clustering based retrieval techniques and also tested for face image retrieval.

Face Match for Family Reunification

Despite the many advances in face recognition technology, practical face detection and matching for unconstrained images remain challenging. A real-world Face Image Retrieval (FIR) system is described in this paper. It is based on optimally weighted image descriptor ensemble utilized in single-image-per-person (SIPP) approach that works with large unconstrained digital photo collections. The described visual search can be deployed in many applications, e.g. person location in post-disaster scenarios, helping families reunite quicker. It provides efficient means for face detection, matching and annotation, working with images of variable quality, requiring no time-consuming training, yet showing commercial performance levels.

Supervised deep hashing for scalable face image retrieval

Hashing has been widely utilized for Approximate Nearest Neighbor (ANN) search due to its fast retrieval speed and low storage cost. In this work, we propose a novel supervised hashing method for scalable face image retrieval, i.e., Deep Hashing based on Classification and Quantization errors (DHCQ), by simultaneously learning feature representations of images, hash codes and classifiers. The supervised information and the deep architecture are collaboratively explored. Specifically, a deep convolutional network is introduced to learn discriminative feature representations, which are directly used to generate hash codes and predict labels of images. The quantization errors and the prediction errors jointly guide the learning of the deep network. They are highly interrelated and promoted each other. It is worth noting that the proposed method is a general hashing method and can be applied to the general image retrieval task. Extensive experiments on two face image datasets and one general image dataset demonstrate the effectiveness of the proposed method compared with several state-of-the-art hashing methods.

Content based Feature Combination Method for Face Image Retrieval using Neural Network and SVM Classifier for Face Recognition

Objectives: To propose a CBIR based face image retrieval and identification model. Method/Analysis: A model of hybrid face recognition system based on CBIR and SVM is proposed. The feature vectors from the face image database are generated by using Gabor wavelet (GW), wavelet Transformation (WT), and Principal Component Analysis (PCA). For the face image retrieval purpose, Artificial Neural Network (ANN) is adopted and its performance on the retrieval process is evaluated with PCA, WT, GW and their fusion as a feature vector. The query image is recognized from the faces returned by retrieval process by using Support Vector Machine (SVM). The experimental results indicate that the fusion of PCA, WT and GW features as a feature vector performs reasonably well for retrieval and recognition process. The experimental results also demonstrate the efficiency of the proposed approach for face recognition over existing methods when considering different performance measures such as system running time, Receiver Operating Characteristic (ROC) curve and recognition accuracy. The proposed model was evaluated with two face databases viz. Unconstrained Facial Image (UFI) and Oracle Research Laboratory (ORL) face database with a recognition accuracy of 95.42% and 98.75% respectively. Finding: The proposed system has been tested for retrieval and recognition and found with reasonable retrieval time and high recognition rate. Novelty/Improvement: The conventional model-based face recognition systems are limited in several aspects, like (1) It is usually time-consuming and expensive to collect a large amount of training facial images, (2) It is usually difficult to generalize the models when new training data or new persons are added, in which an intensive retraining process is usually required and (3) The recognition performance often scales poorly when the number of persons/ classes are very large. The proposed CBIR based retrieval and the recognition system is intended to take care of the above limitation.

Subspace mapping of facial attributes for face image retrieval

Subspace mapping of facial attributes for face image retrieval

Face image retrieval: super-resolution based on sketch-photo transformation

Considering the crucial role of face image in modern intelligent system, face image retrieval has attracted more attention for authentication, surveillance, law enforcement, and security control. In most cases, we cannot obtain the suspect’s face image directly and the best substitute is a face sketch of criminal suspect drawn by artist according to eyewitness description. It is a key step in the criminal investigation process to narrow down criminal suspect using the face sketch. At first, the face sketch is transformed to a pseudo-photo for subsequent utilization. Transformation is performed according to the classic eigenface algorithm and enhanced by super-resolution. Matching between reconstructed pseudo-photo and real face photographs is performed by Hash encoding and iterative quantization. We carried out our ideas on two public face databases, and the sketch face images are generated by photo-shopping software program. The experimental results show that the corresponding face images can be retrieved according to the input face sketch and super-resolution can effectively enhance the image quality and detail information of the pseudo-photo. Hash encoding and iterative quantization achieve the quick search of approximate face images.

A retrieval algorithm for specific face images in airport surveillance multimedia videos on cloud computing platform

During retrieval process for specific face images in the airport surveillance multimedia video on cloud computing platform, because pattern of faces are complicated and vulnerable to interference, when the traditional algorithm is used for face retrieval, the accuracy and efficiency are reduced and robustness is low. A new retrieval algorithm for specific face images is proposed, and this method is deployed to the cloud computing platform. Harr face cascade classifier is used to detect face images in airport surveillance multimedia video, in order to find out the missing face in airport surveillance multimedia video, the block matching method is introduced for face tracking, and the missing face in the video is obtained. PCA method is utilized to extract specific facial features, and discriminant analysis method is used to compare the extracted feature information with the specific face, so as to realize the specific face image retrieval. Experimental results show that the proposed algorithm has a high retrieval efficiency and precision.

Structured detail enhancement for cross-modality face synthesis

Transformation, matching, and retrieval of face images in different modalities are often encountered in many applications. Cross-modality face synthesis, which aims to transform face images between modalities, can serve as an intermediate representation for these tasks, and remains an active yet challenging topic in low level vision. Most existing cross-modality face synthesis methods are effective in recovering global features for a given photo, but fail to capture fine-scale details or introduce some extra noise in the synthesis results. In this paper, we propose a two-step algorithm to tackle this problem by enhancing structured details after initial synthesis. In the first step, a combination of quasi-smooth area synthesis and facial component synthesis is proposed to generate more plausible facial components. In the second step, an MAP-based model is constructed to introduce more facial details into the initial synthesis, meanwhile eliminate noise. Experiments on face sketch synthesis and near-infrared face synthesis demonstrate the effectiveness of the proposed method in enhancing fine-scale facial details while reducing noise.

Implementing Kernel Methods Incrementally by Incremental Nonlinear Projection Trick.

Recently, the nonlinear projection trick (NPT) was introduced enabling direct computation of coordinates of samples in a reproducing kernel Hilbert space. With NPT, any machine learning algorithm can be extended to a kernel version without relying on the so called kernel trick. However, NPT is inherently difficult to be implemented incrementally because an ever increasing kernel matrix should be treated as additional training samples are introduced. In this paper, an incremental version of the NPT (INPT) is proposed based on the observation that the centerization step in NPT is unnecessary. Because the proposed INPT does not change the coordinates of the old data, the coordinates obtained by INPT can directly be used in any incremental methods to implement a kernel version of the incremental methods. The effectiveness of the INPT is shown by applying it to implement incremental versions of kernel methods such as, kernel singular value decomposition, kernel principal component analysis, and kernel discriminant analysis which are utilized for problems of kernel matrix reconstruction, letter classification, and face image retrieval, respectively.

Multi Model based Biometric Image Retrieval for Enhancing Security

Objective: Biometric image retrieval becomes the most popular application scenario in the real world where the many application are started to use bio metric authentication. It plays a key role in the security management system of the real world application to prevent the malicious user access. Storage and management of the bio metric details of multiple users are the biggest issue which may lead to inefficient performance. In our previous work, two model bio metric based human recognition is introduced where the iris and finger prints are considered. But it may not recognize the human system in case of missing information and the corruption of extracted features. To overcome this problem in this work multi model based human recognition is introduced. Methodology: The four models considered in this work are: Iris, finger print, face and the palm print. The unique methodology is used for extracting the features from these models. In the previous research work, Speeded Up Robust Features (SURF) algorithm is used for iris feature extraction and Improved Locality-Sensitive Hashing (ILSH) indexing method is used for finger print feature extraction. In our proposed methodology, weighted attributes sparse code words indexing is proposed for face image retrieval where the attribute scores are selected optimally by using the BAT algorithm and a Novel hierarchical Indexing approach is proposed for Palm Print Image Retrieval. Finally these extracted features are fused together to generate a single model from the multiple models by using the approach called uni-model indexing scheme. Findings: The experimental tests conducted prove that the proposed methodology provides better result than the existing approach in terms of improved retrieval accuracy. This simulation evaluation is done in the MATLAB simulation environment for the both existing and proposed approach which is then compared with each other to obtain the performance improvement. Applications: The findings of this work prove that the proposed approach provides better result than the existing approach in terms of improved accuracy. This approach can be applied in the many research field where the security is required for preventing from the intruders attack like military, hospital management etc.

Making People's Life Much Easier by Effective Face Image Retrieval from the Popular Online Social Network such as Facebook

The primary objective of this paper is to make people's life much easier by retrieving images containing faces of the same person appearing in the query image, from the popular online social network like Facebook (web-scale image database) containing millions of face images effectively and accurately. In this work face image of a person will be used as a query image to retrieve images of the same person with various poses, expressions, and illumination. Such face retrieval system has many applications, including name-based face image search, face tagging in images and videos, copyright enforcement, labeling faces in photos from the popular online social network such as Facebook etc. Also, this paper proposes a method for Automatic Image Registration in which pair of images with differences in rotation, translation, and spectral content is registered for remote sensing applications using multisensors. This work identifies that an image edge feature extraction using perceptional organization methods based on the rules of human perception are allowed us to reduce noise and over segmentation problem in image feature extraction. Generic edge tokens are perceptually significant edge features which are descriptive in the representation and therefore can be manipulated qualitatively in supporting perceptional grouping. This paper provided a new technique that can speed up registration (by reducing data) and improve accuracy by selecting a subimage for the registration process based on its ability to provide accurate results and determined a faster method for implementing multi-resolution image registration. The proposed work not only reduces the computation cost of future work, but it also greatly reduces the complexity and confusion caused by over-segmentation.

Discriminative Feature Extraction by a Neural Implementation of Canonical Correlation Analysis.

The canonical correlation analysis (CCA) aims at measuring linear relationships between two sets of variables (views) that can be used for feature extraction in classification problems with multiview data. However, the correlated features extracted by the CCA may not be class discriminative, since CCA does not utilize the class labels in its traditional formulation. Although there is a method called discriminative CCA (DCCA) that aims to increase the discriminative ability of CCA inspired from the linear discriminant analysis (LDA), it has been shown that the extracted features with this method are identical to those by the LDA with respect to an orthogonal transformation. Therefore, DCCA is simply equivalent to applying single-view (regular) LDA to each one of the views separately. Besides, DCCA and the other similar DCCA approaches have generalization problems due to the sample covariance matrices used in their computation, which are sensitive to outliers and noisy samples. In this paper, we propose a method, called discriminative alternating regression (D-AR), to explore correlated and also discriminative features. D-AR utilizes two (alternating) multilayer perceptrons, each with a linear hidden layer, learning to predict both the class labels and the outputs of each other. We show that the features found by D-AR on training sets significantly accomplish higher classification accuracies on test sets of facial expression recognition, object recognition, and image retrieval experimental data sets.

Novel histograms kernels with structural properties

This paper introduces a new similarity measure for comparing histograms, named Weighted Distribution Matching, which bases the comparison not only in the specific bin values but also in the shape of the histograms. It is proved that the proposed similarity is a conditionally positive definite kernel. The space where histograms lie is studied, and some intuitively desirable structural properties are introduced. The most representative measures of the state of art were compared on the basis of these properties. Experiments conducted on face recognition and image retrieval validate the proposal.

Center symmetric local binary co-occurrence pattern for texture, face and bio-medical image retrieval

Content based image retrieval is a common problem for a large image database. Many methods have been proposed for image retrieval for some particular type of datasets. In the proposed work, a new image retrieval technique has been introduced. This technique is useful for different kind of dataset. In the proposed method, center symmetric local binary pattern has been extracted from the original image to obtain the local information. Co-occurrence of pixel pairs in local pattern map have been observed in different directions and distances using gray level co-occurrence matrix. Earlier methods have utilized histogram to extract the frequency information of local pattern map but co-occurrence of pixel pairs is more robust than frequency of patterns. The proposed method is tested on three different category of images, i.e., texture, face and medical image database and compared with typical state-of-the-art local patterns.

Scalable attribute-driven face image retrieval

In recent years an explosion of online multimedia data has been witnessed. As an example, abundant photos recording every aspect of human life are available through social media. Among tremendous amount of photos, a significant fraction contains human faces. Faces are usually salient features of the photos. To understand and extract useful information from such gigantic data corpus, efficient and effective retrieval algorithms are demanded. Most face retrieval techniques rely on low-level image features to compare faces based on visual similarity. However, as humans we tend to simplify the recognition task by utilizing human attributes such as gender or race to help differentiate people on a higher semantic level. In this paper, we propose to use human attributes as high-level semantic cues to determine people׳s identities. To this end, we develop discriminative image features with attribute information encoded to achieve more accurate face image retrieval. To guarantee scalability, we propose using a binary coding scheme for the proposed attributed-based features. A re-ranking step after initial retrieval is incorporated to further improve the retrieval performance. We demonstrate the superiority of the proposed method compared to state-of-the-art on the LFW and Pubfig face datasets.

Visually Interpreting Names as Demographic Attributes by Exploiting Click-Through Data

Name of an identity is strongly influenced by his/her cultural background such as gender and ethnicity, both vital attributes for user profiling, attribute-based retrieval, etc. Typically, the associations between names and attributes (e.g., people named "Amy" are mostly females) are annotated manually or provided by the census data of governments. We propose to associate a name and its likely demographic attributes by exploiting click-throughs between name queries and images with automatically detected facial attributes. This is the first work attempting to translate an abstract name to demographic attributes in visual-data-driven manner, and it is adaptive to incremental data, more countries and even unseen names (the names out of click-through data) without additional manual labels. In the experiments, the automatic name-attribute associations can help gender inference with competitive accuracy by using manual labeling. It also benefits profiling social media users and keyword-based face image retrieval, especially for contributing 12% relative improvement of accuracy in adapting to unseen names.

Partially Occluded Facial Image Retrieval Based on a Similarity Measurement

We present a partially occluded facial image retrieval method based on a similarity measurement for forensic applications. The main novelty of this method compared with other occluded face recognition algorithms is measuring the similarity based on Scale Invariant Feature Transform (SIFT) matching between normal gallery images and occluded probe images. The proposed method consists of four steps: (i) a Self-Quotient Image (SQI) is applied to input images, (ii) Gabor-Local Binary Pattern (Gabor-LBP) histogram features are extracted from the SQI images, (iii) the similarity between two compared images is measured by using the SIFT matching algorithm, and (iv) histogram intersection is performed on the SIFT-based similarity measurement. In experiments, we have successfully evaluated the performance of the proposed method with the commonly used benchmark database, including occluded facial images. The results show that the correct retrieval ratio was 94.07% in sunglasses occlusion and 93.33% in scarf occlusion. As such, the proposed method achieved better performance than other Gabor-LBP histogram-based face recognition algorithms in eyes-hidden occlusion of facial images.

Semantic-Based Facial Image-Retrieval System with Aid of Adaptive Particle Swarm Optimization and Squared Euclidian Distance

The semantic-based facial image-retrieval system is concerned with the process of retrieving facial images based on the semantic information of query images and database images. The image-retrieval systems discussed in the literature have some drawbacks that degrade the performance of facial image retrieval. To reduce the drawbacks in the existing techniques, we propose an efficient semantic-based facial image-retrieval (SFIR) system using APSO and squared Euclidian distance (SED). The proposed technique consists of three stages: feature extraction, optimization, and image retrieval. Initially, the features are extracted from the database images. Low-level features (shape, color, and texture) and high-level features (face, mouth, nose, left eye, and right eye) are the two features used in the feature-extraction process. In the second stage, a semantic gap between these features is reduced by a well-known adaptive particle swarm optimization (APSO) technique. Afterward, a squared Euclidian distance (SED) measure will be utilized to retrieve the face images that have less distance with the query image. The proposed semantic-based facial image-retrieval (SFIR) system with APSO-SED will be implemented in working platform of MATLAB, and the performance will be analyzed.

Face Recognition System Based on Sparse Codeword Analysis

In recent times, large-scale content-based face image retrieval has grown up with rapid improvement and it is an enabling technology for many emerging applications. Content based face image retrieval is done by computing the similarity between images in the databases and the input/query face image. Content based face image retrieval systems retrieves the image only using low level features therefore the retrieval rate is low in this system. To improve the retrieval rate sparse codeword based scalable face image retrieval system is developed. This system uses both low level features and high level human attributes. The proposed system has several stages to retrieve the images; 1. Low level features are extracted using LTP descriptor and utilize the automatically detected high level human attributes such as hair, Gender and race. 2. Sparse codeword techniques are applied on the low level features and attributes to generate the codeword. 3. The third stage is an indexing; in the indexing attribute embedded inverted indexing method is used. Using the methods mentioned above, face image retrieval system has achieved promising retrieval result. Experiment is conducted on different dataset such as pub fig, LFW and FERET. Among those dataset LFW dataset achieve higher performance.

Distributed Face Recognition in Wireless Sensor Networks

As one of the most successful applications of image analysis and understanding, face recognition has recently received significant attention, especially during the past few years. In order to construct an autonomous and robust biometric security system, this paper explores the application of face recognition technique in wireless sensor networks. Given the limited technological resources of sensor nodes, new challenges remain to be met. In this work, a facial component-based recognition mechanism is firstly applied to ensure the recognition accuracy. Secondly, in order to address the problem of resource constraints, a distributed scheme based on K-d trees is deployed for both the face image transmission and retrieval. According to the simulation results, the proposed method is capable of achieving considerable energy efficiency, while assuring the recognition accuracy.