A survey on heterogeneous face recognition
TL;DR: This survey provides a comprehensive review of established techniques and recent developments in HFR, and offers a detailed account of datasets and benchmarks commonly used for evaluation.
About: This article is published in Image and Vision Computing.The article was published on 2016-12-01 and is currently open access. It has received 114 citations till now.
Citations
More filters
TL;DR: A comprehensive review of the recent developments on deep face recognition can be found in this paper, covering broad topics on algorithm designs, databases, protocols, and application scenes, as well as the technical challenges and several promising directions.
353 citations
TL;DR: Major deep learning concepts pertinent to face image analysis and face recognition are reviewed, and a concise overview of studies on specific face recognition problems is provided, such as handling variations in pose, age, illumination, expression, and heterogeneous face matching.
312 citations
Posted Content•
TL;DR: A comprehensive review of pose-invariant face recognition methods can be found in this paper, where pose-robust feature extraction approaches, multi-view subspace learning approaches, face synthesis approaches, and hybrid approaches are compared.
Abstract: The capacity to recognize faces under varied poses is a fundamental human ability that presents a unique challenge for computer vision systems. Compared to frontal face recognition, which has been intensively studied and has gradually matured in the past few decades, pose-invariant face recognition (PIFR) remains a largely unsolved problem. However, PIFR is crucial to realizing the full potential of face recognition for real-world applications, since face recognition is intrinsically a passive biometric technology for recognizing uncooperative subjects. In this paper, we discuss the inherent difficulties in PIFR and present a comprehensive review of established techniques. Existing PIFR methods can be grouped into four categories, i.e., pose-robust feature extraction approaches, multi-view subspace learning approaches, face synthesis approaches, and hybrid approaches. The motivations, strategies, pros/cons, and performance of representative approaches are described and compared. Moreover, promising directions for future research are discussed.
263 citations
13 Jun 2016
TL;DR: A deep TransfeR NIR-VIS heterogeneous face recognition neTwork (TRIVET) with deep convolutional neural network with ordinal measures to learn discriminative models achieves state-of-the-art recognition performance on the most challenging CASIA Nir-VIS 2.0 Face Database.
Abstract: One task of heterogeneous face recognition is to match a near infrared (NIR) face image to a visible light (VIS) image. In practice, there are often a few pairwise NIR-VIS face images but it is easy to collect lots of VIS face images. Therefore, how to use these unpaired VIS images to improve the NIR-VIS recognition accuracy is an ongoing issue. This paper presents a deep TransfeR NIR-VIS heterogeneous facE recognition neTwork (TRIVET) for NIR-VIS face recognition. First, to utilize large numbers of unpaired VIS face images, we employ the deep convolutional neural network (CNN) with ordinal measures to learn discriminative models. The ordinal activation function (Max-Feature-Map) is used to select discriminative features and make the models robust and lighten. Second, we transfer these models to NIR-VIS domain by fine-tuning with two types of NIR-VIS triplet loss. The triplet loss not only reduces intra-class NIR-VIS variations but also augments the number of positive training sample pairs. It makes fine-tuning deep models on a small dataset possible. The proposed method achieves state-of-the-art recognition performance on the most challenging CASIA NIR-VIS 2.0 Face Database. It achieves a new record on rank-1 accuracy of 95.74% and verification rate of 91.03% at FAR=0.001. It cuts the error rate in comparison with the best accuracy [27] by 69%.
128 citations
Cites methods from "A survey on heterogeneous face reco..."
...Traditional methods on heterogeneous face recognition mainly focus on three strategies to alleviate the crossmodal gap [13]: designing invariant features for different modalities, transforming one face modality to the other, and projecting both image modalities onto a common subspace....
[...]
TL;DR: In this article, a method to generate very large training data sets of synthetic images by compositing real face images in a given data set is proposed, which enables to learn models from as few as 10, 000 training images, which perform on par with models trained from 500, 000 images.
Abstract: Deep convolutional neural networks have recently proven extremely effective for difficult face recognition problems in uncontrolled settings. To train such networks, very large training sets are needed with millions of labeled images. For some applications, such as near-infrared (NIR) face recognition, such large training data sets are not publicly available and difficult to collect. In this paper, we propose a method to generate very large training data sets of synthetic images by compositing real face images in a given data set. We show that this method enables to learn models from as few as 10 000 training images, which perform on par with models trained from 500 000 images. Using our approach, we also obtain state-of-the-art results on the CASIA NIR-VIS2.0 heterogeneous face recognition data set.
112 citations
References
More filters
Proceedings Article•
03 Dec 2012TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.
73,978 citations
Book•
01 Sep 1988
TL;DR: In this article, the authors present the computer techniques, mathematical tools, and research results that will enable both students and practitioners to apply genetic algorithms to problems in many fields, including computer programming and mathematics.
Abstract: From the Publisher:
This book brings together - in an informal and tutorial fashion - the computer techniques, mathematical tools, and research results that will enable both students and practitioners to apply genetic algorithms to problems in many fields
Major concepts are illustrated with running examples, and major algorithms are illustrated by Pascal computer programs No prior knowledge of GAs or genetics is assumed, and only a minimum of computer programming and mathematics background is required
52,797 citations
20 Jun 2009
TL;DR: A new database called “ImageNet” is introduced, a large-scale ontology of images built upon the backbone of the WordNet structure, much larger in scale and diversity and much more accurate than the current image datasets.
Abstract: The explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data. But exactly how such data can be harnessed and organized remains a critical problem. We introduce here a new database called “ImageNet”, a large-scale ontology of images built upon the backbone of the WordNet structure. ImageNet aims to populate the majority of the 80,000 synsets of WordNet with an average of 500-1000 clean and full resolution images. This will result in tens of millions of annotated images organized by the semantic hierarchy of WordNet. This paper offers a detailed analysis of ImageNet in its current state: 12 subtrees with 5247 synsets and 3.2 million images in total. We show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets. Constructing such a large-scale database is a challenging task. We describe the data collection scheme with Amazon Mechanical Turk. Lastly, we illustrate the usefulness of ImageNet through three simple applications in object recognition, image classification and automatic object clustering. We hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.
49,639 citations
"A survey on heterogeneous face reco..." refers background in this paper
...sketch [25, 32, 25, 39], are much smaller than those used in homogeneous face recognition [82] and broader computer vision [150] problems....
[...]
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 large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.
Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0%, respectively, which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overfitting in the fully connected layers we employed a recently developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.
33,301 citations
Additional excerpts
...effective feature representations [154]....
[...]