NIR Face Research Articles

Automated anomalous child repetitive head movement identification through transformer networks.

The increasing prevalence of behavioral disorders in children is of growing concern within the medical community. Recognising the significance of early identification and intervention for atypical behaviors, there is a consensus on their pivotal role in improving outcomes. Due to inadequate facilities and a shortage of medical professionals with specialized expertise, traditional diagnostic methods have been unable to effectively address the rising incidence of behavioral disorders. Hence, there is a need to develop automated approaches for the diagnosis of behavioral disorders in children, to overcome the challenges with traditional methods. The purpose of this study is to develop an automated model capable of analyzing videos to differentiate between typical and atypical repetitive head movements in. To address problems resulting from the limited availability of child datasets, various learning methods are employed to mitigate these issues. In this work, we present a fusion of transformer networks, and Non-deterministic Finite Automata (NFA) techniques, which classify repetitive head movements of a child as typical or atypical based on an analysis of gender, age, and type of repetitive head movement, along with count, duration, and frequency of each repetitive head movement. Experimentation was carried out with different transfer learning methods to enhance the performance of the model. The experimental results on five datasets: NIR face dataset, Bosphorus 3D face dataset, ASD dataset, SSBD dataset, and the Head Movements in the Wild dataset, indicate that our proposed model has outperformed many state-of-the-art frameworks when distinguishing typical and atypical repetitive head movements in children.

Facial feature embedded CycleGAN for VIS–NIR translation

Visible and near-infrared (VIS–NIR) heterogeneous face recognition remains a challenging task due to distinctions between spectral components of two modalities and insufficiently pairwise VIS–NIR data. Inspired by the cycle-consistent generative adversarial network (CycleGAN), this paper proposes a facial feature embedded CycleGAN to translate between VIS and NIR face images, aiming to enable the distributions of translated (fake) images to be similar as those of true images. To learn the particular feature of NIR or VIS domain while preserving common facial representation between VIS and NIR domains, a facial feature extractor (FFE), tailored specifically for extracting effective feature from face images, is embedded in the generator of original CycleGAN. For implementing the FFE, we use the MobileFaceNet which is pre-trained on a VIS face database. The domain-invariant feature learning is enhanced by proposing a new pixel consistency loss. Additionally, we establish a new WHU VIS–NIR database including varies in face rotation and expressions to enrich the insufficient training data. Moreover, experiments on the well-known Oulu-CASIA NIR–VIS database and our WHU VIS–NIR database validate the potential benefit of the proposed FFE-based CycleGAN (FFE-CycleGAN). In particular, we achieve 96.5% accuracy on Oulu-CASIA and 98.9% accuracy on WHU VIS–NIR.

Research of ultra-light InGaAs NIR face detection algorithm

Research of ultra-light InGaAs NIR face detection algorithm

Outside Box and Contactless Palm Vein Recognition Based on a Wavelet Denoising ResNet

Palm vein recognition is a high-security biometric. Outside the NIR capture box and contactless palm vein recognition are more popular but challenging. The users feel comfortable outside the NIR capture box but face more optical blurring brought by visible light. Contactless capture gestures solve the hygienic problem but face the image rotation, position translation, and scale transformation which makes classification difficult especially in large-scale databases. To address these problems, we develop a wavelet denoising ResNet, which consists of two models: the wavelet denoising (WD) model and the squeeze-and-excitation ResNet18 (SER) model. The WD model focuses on removing noise from skin scattering and optical blurring from palm vein images. The WD model enhances the low-frequency feature into a deep learning feature by residual learning technology. This strategy increases the weight of an effective handcrafted feature in the deep learning network. The SER model overcomes rotation, position translation, and scale transformation by selectively emphasizing classification features and weakening less useful features. To train and verify the network, an inside box palm vein image database and an outside box palm vein image database are set up. The Tongji contactless palm vein image database was also employed in the experiments. The validity and superiority of our network are verified in a series of experiments.

Partial NIR-VIS Heterogeneous Face Recognition With Automatic Saliency Search

Near-infrared-visual (NIR-VIS) heterogeneous face recognition (HFR) aims to match NIR face images with the corresponding VIS ones. It is a challenging task due to the sensing gaps among different modalities. Occlusions in the input face images make the task extremely complex. To tackle these problems, we present a Saliency Search Network (SSN) to extract domain-invariant identity features. We propose to automatically search the efficient parts of face images in a modality-aware manner, and remove redundant information. Moreover, the searching process is guided by an information bottleneck network, which mitigates the overfitting problems caused by small datasets. Extensive experiments on both complete and partial NIR-VIS HFR on multiple datasets demonstrate the effectiveness and robustness of the proposed method to modality discrepancy and occlusions.

Adversarial Cross-Spectral Face Completion for NIR-VIS Face Recognition

Near infrared-visible (NIR-VIS) heterogeneous face recognition refers to the process of matching NIR to VIS face images. Current heterogeneous methods try to extend VIS face recognition methods to the NIR spectrum by synthesizing VIS images from NIR images. However, due to the self-occlusion and sensing gap, NIR face images lose some visible lighting contents so that they are always incomplete compared to VIS face images. This paper models high-resolution heterogeneous face synthesis as a complementary combination of two components: a texture inpainting component and a pose correction component. The inpainting component synthesizes and inpaints VIS image textures from NIR image textures. The correction component maps any pose in NIR images to a frontal pose in VIS images, resulting in paired NIR and VIS textures. A warping procedure is developed to integrate the two components into an end-to-end deep network. A fine-grained discriminator and a wavelet-based discriminator are designed to improve visual quality. A novel 3D-based pose correction loss, two adversarial losses, and a pixel loss are imposed to ensure synthesis results. We demonstrate that by attaching the correction component, we can simplify heterogeneous face synthesis from one-to-many unpaired image translation to one-to-one paired image translation, and minimize the spectral and pose discrepancy during heterogeneous recognition. Extensive experimental results show that our network not only generates high-resolution VIS face images but also facilitates the accuracy improvement of heterogeneous face recognition.

Face Recognition in Thermal Images based on Sparse Classifier

Despite recent advances in face recognition systems, they suffer from serious problems because of the extensive types of changes in human face (changes like light, glasses, head tilt, different emotional modes). Each one of these factors can significantly reduce the face recognition accuracy. Several methods have been proposed by researchers to overcome these problems. Nonetheless, in recent years, using thermal images has gain more attention among the introduced solutions as an effective and unique solution. This article studies the performance of sparse processing techniques when facing with challenges of face recognition problem in thermal images. Also, the potential of the sparse classifier algorithm to receive information directly from input images without using any feature extraction algorithms was studied. The obtained results indicated that the sparse processing techniques outperform the Eigenface and KNN algorithms in terms of addressing the challenges of thermal images. In this work, USTC NVIN and CBSR NIR face datasets were used for simulation purposes. These datasets include the images with different emotional states (sad, happy, etc.) captured in different light conditions; also the images are captured both with and without wearing glasses. Simulation results have shown that sparse classifier can be an effective algorithm for the face recognition problem in thermal images.

Local SVD based NIR face retrieval

From last decade, local descriptor such as Local Binary Pattern (LBP) is accepted as a very prominent feature descriptor for characterizing the images such as faces. The performance of such descriptors depends upon the local relationship of the image. The local relationship of the image can be utilized in more discriminative and robust way after some preprocessing as compared to the original image. The preprocessed images in the form of 4 sub-bands (i.e. S, U, V, and D sub-bands) are obtained by applying the Singular Value Decomposition (SVD) over the original image. The local descriptors are computed over these sub-bands (mainly S sub-band) and termed as the SVD based local descriptors. The performance of four local descriptors over SVD sub-bands are tested for near-infrared face retrieval using PolyU-NIR and CASIA-NIR face databases, and compared with the results obtained using descriptors without SVD sub-band. The experimental results confirm the superiority of using S sub-band of SVD in terms of performance of the local descriptors over NIR face databases.

Kernel-aligned multi-view canonical correlation analysis for image recognition

Existing kernel-based correlation analysis methods mainly adopt a single kernel in each view. However, only a single kernel is usually insufficient to characterize nonlinear distribution information of a view. To solve the problem, we transform each original feature vector into a 2-dimensional feature matrix by means of kernel alignment, and then propose a novel kernel-aligned multi-view canonical correlation analysis (KAMCCA) method on the basis of the feature matrices. Our proposed method can simultaneously employ multiple kernels to better capture the nonlinear distribution information of each view, so that correlation features learned by KAMCCA can have well discriminating power in real-world image recognition. Extensive experiments are designed on five real-world image datasets, including NIR face images, thermal face images, visible face images, handwritten digit images, and object images. Promising experimental results on the datasets have manifested the effectiveness of our proposed method.

Overview of principal component analysis algorithm

Principal component analysis (PCA) algorithm has been extensively employed in face recognition. However, existing PCA algorithms have some limitations in face recognition. In order to overcome these limitations, researchers proposed some extended-PCA algorithms. In this paper, we present algorithm implementation of the original PCA algorithm and main extended-PCA algorithms including two-dimensional PCA (2DPCA), 2DPCA-based feature fusion approach, the kernel PCA (KPCA), the modular PCA, improved KPCA (IKPCA), efficient sparse KPCA (ESKPCA) and incremental PCA (IPCA). 2DPCA directly computes the projection of a matrix onto a transforming axis. 2DPCA-based feature fusion approach combines the features generated from the two schemes of 2DPCA. KPCA can perform well in extracting features from samples whose components have nonlinear relations. The modular PCA approach divides the original face image into sub-images and applies the original PCA approach to each of these sub-images. The IKPCA algorithm improves KPCA for more efficient feature extraction. The efficient sparse KPCA (ESKPCA) improves the computational efficiency of the previous sparse KPCA methods on large-scale training sample sets. Incremental PCA (IPCA) overcomes the limitation which is hard to scale up the developed systems. In order to compare the recognition rate of these algorithms in face recognition, a series of experiments are performed on three face image databases: ORL, Yale and NIR face databases.

Near infrared face recognition using Zernike moments and Hermite kernels

This work proposes a novel face recognition method based on Zernike moments (ZMs) and Hermite kernels (HKs) to cope with variations in facial expression, changes in head pose and scale, occlusions due to wearing eyeglasses and the effects of time lapse. Near infrared images are used to tackle the impact of illumination changes on face recognition, and a combination of global and local features is utilized in the decision fusion step. In the global part, ZMs are used as a feature extractor and in the local part, the images are partitioned into multiple patches and filtered patch-wise with HKs. Finally, principal component analysis followed by linear discriminant analysis is applied to data vectors to generate salient features and decision fusion is applied on the feature vectors to properly combine both global and local features. Experimental results on CASIA NIR and PolyU NIR face databases clearly show that the proposed method achieves significantly higher face recognition accuracy compared with existing methods.

Matching NIR Face to VIS Face Using Transduction

Visual versus near infrared (VIS-NIR) face image matching uses an NIR face image as the probe and conventional VIS face images as enrollment. It takes advantage of the NIR face technology in tackling illumination changes and low-light condition and can cater for more applications where the enrollment is done using VIS face images such as ID card photos. Existing VIS-NIR techniques assume that during classifier learning, the VIS images of each target people have their NIR counterparts. However, since corresponding VIS-NIR image pairs of the same people are not always available, which is often the case, so those methods cannot be applied. To address this problem, we propose a transductive method named transductive heterogeneous face matching (THFM) to adapt the VIS-NIR matching learned from training with available image pairs to all people in the target set. In addition, we propose a simple feature representation for effective VIS-NIR matching, which can be computed in three steps, namely Log-DoG filtering, local encoding, and uniform feature normalization, to reduce heterogeneities between VIS and NIR images. The transduction approach can reduce the domain difference due to heterogeneous data and learn the discriminative model for target people simultaneously. To the best of our knowledge, it is the first attempt to formulate the VIS-NIR matching using transduction to address the generalization problem for matching. Experimental results validate the effectiveness of our proposed method on the heterogeneous face biometric databases.