Figure 2 - uploaded by Zhengyou Zhang
Content may be subject to copyright.
Illustration of the integral image and Haar-like rectangle features (a-f).  

Illustration of the integral image and Haar-like rectangle features (a-f).  

Source publication
Book
Full-text available
Face detection has been one of the most studied topics in the computer vision literature. In this technical report, we survey the recent advances in face detection for the past decade. The seminal Viola-Jones face detector is first reviewed. We then survey the various techniques according to how they extract features and what learning algorithms ar...

Contexts in source publication

Context 1
... ii(x, y) is the integral image at pixel location (x, y) and i(x , y ) is the original image. Using the integral image to compute the sum of any rectangular area is extremely efficient, as shown in Fig. 2. The sum of pixels in rectangle region ABCD can be calculated as: ...
Context 2
... only requires four array references. The integral image can be used to compute simple Haar- like rectangular features, as shown in Fig. 2 (a-f). The fea- tures are defined as the (weighted) intensity difference be- tween two to four rectangles. For instance, in feature (a), the feature value is the difference in average pixel value in the gray and white rectangles. Since the rectangles share corners, the computation of two rectangle features (a and b) requires six array ...
Context 3
... F t (x) is given. For this purpose, we assume the base function pool {f (x)} is in the form of confidence rated decision stumps. That is, a certain form of real feature value h(x) is first ex- tracted from x, h : X → R. For instance, in the Viola-Jones face detector, h(x) is the Haar-like features computed with integral image, as was shown in Fig. 2 (a-f). A decision threshold H divide the output of h(x) into two subregions, u 1 and u 2 , u 1 ∪ u 2 = R. The base function f (x) is ...
Context 4
... Haar-like rectangular features as in Fig. 2 (a-f) are very efficient to compute due to the integral image tech- nique, and provide good performance for building frontal face detectors. In a number of follow-up works, researchers extended the straightforward features with more variations in the ways rectangle features are ...

Similar publications

Article
Full-text available
Facial attribute analysis in the real world scenario is very challenging mainly because of complex face variations. Existing works of analyzing face attributes are mostly based on the cropped and aligned face images. However, this result in the capability of attribute prediction heavily relies on the preprocessing of face detector. To address this...
Conference Paper
Full-text available
This article deals with facial detection and tracking algorithm development. The most efficient method for facial monitoring, namely, template matching technique, was found by considering different tracking techniques. The authors enumerate main drawbacks of the template tracking and improve the algorithm according to the problem set. The following...

Citations

... hertasecurity.com). Firstly, a generic face detector optimized for unconstrained video surveillance scenarios was used to obtain the locations of all faces in each image (Zhang and Zhang, 2010). The threshold was adjusted to find all targeted faces in each photo, and a subsequent manual exploration was conducted to ensure that no false positives were included. ...
Article
Full-text available
The human face is one of the most visible features of our unique identity as individuals. Interestingly, monozygotic twins share almost identical facial traits and the same DNA sequence but could exhibit differences in other biometrical parameters. The expansion of the world wide web and the possibility to exchange pictures of humans across the planet has increased the number of people identified online as virtual twins or doubles that are not family related. Herein, we have characterized in detail a set of “look-alike” humans, defined by facial recognition algorithms, for their multiomics landscape. We report that these individuals share similar genotypes and differ in their DNA methylation and microbiome landscape. These results not only provide insights about the genetics that determine our face but also might have implications for the establishment of other human anthropometric properties and even personality characteristics.
... The second point was enhancing the boosting algorithm using various edited versions of the AdaBoost such as RealBoost [6], LogitBoost [45,74] and KLBoosting [39]. In 2002, Viola and Jones [6] extended their work [63] to resolve the problem of the asymmetry that existed between both negative and positive data sets. ...
... While Viola and Jones [63] used the Haar-like feature in their work, here the authors used joint Haar-like features to extract the structural characteristic of the human face, thereby building more effective weak classifier. Instead of using the Adaboost algorithm, the authors used LogitBoost [45,74] to train the classifier by stage wise selection. Through their experiment, the authors achieved better results using their boosting algorithm compared to the Adaboost [65,60]. ...
Article
Full-text available
Real time face detection techniques are needed in a wide range of fields. Therefore, developing a high, accurate and efficient near-real-time face detection method has become a major concern for both industrial and research communities. This paper introduces a critical comparison to a variety of face-detection methods, namely, (1) Haar-like cascade, (2) Linear Binary Pattern cascade (LBP), (3) Histogram of Oriented Gradients with Support Vector Machine (HOG) and (4) Convolutional Neural Network based algorithms (CNN) using video sequences rather than static images. Different experiments were conducted to evaluate the performance of these techniques on constraint and spontaneous video sequences from the database for Remote Collaborative and Affective Interactions (RECOLA) and the Database for Emotion Analysis using Physiological Signals (DEAP). The experimental results show that CNN based algorithm is more efficient compared to other approaches. It achieves an average detection rate of 99.99% for the DEAP database and 84.23% for the RECOLA database. However, it is the slowest when it comes to detecting faces with an average number of frames per second (FPS) of 2.12 and 2.58. Meanwhile LBP method is the fastest among the proposed methods with an average FPS of 25.58 and 33.79.
... Using facial features, animal monitoring does not require a direct contact with the sensor and the animal will thus be at ease. Animal face detection can be used for many security applications that identify animal faces, gender/age detection and visual monitoring [49]. Ensuring safety for animals is an important task, particularly for breeders in many commercial fields such as horse race, livestock buying and selling as well as cow breeding. ...
Article
Full-text available
This paper focuses on the face detection problem of three popular animal categories that need control such as horses, cats and dogs. Existing detectors are generally based on Convolutional Neural Networks (CNNs) as backbones. CNNs are strong and fascinating classification tools but present some weak points such as the big number of layers and parameters, require a huge dataset and ignore the relationship between image parts. To be precise, to deal with these problems, this paper contributes to present a new Convolutional Neural Network for Animal Face Detection (CNNAFD), a new backbone CNNAFD-MobileNetV2 for animal face detection and a new Tunisian Horse Detection Database (THDD). CNNAFD used a processed filters based on gradient features and applied with a new way. A new sparse convolutional layer ANOFS-Conv is proposed through a sparse feature selection method known as Automated Negotiation-based Online Feature Selection (ANOFS). The ANOFS method is used as a training optimizer for the new ANOFS-Conv layer. CNNAFD ends by stacked fully connected layers which represent a strong classifier. The fusion of CNNAFD and MobileNetV2 constructs the new network CNNAFD-MobileNetV2 which improves the classification results and gives better detection decisions. The proposed detector with the new CNNAFD-MobileNetV2 network provides effective results and proves to be competitive with the detectors of the related works with an Average Precision equal to 98.28%, 99.78%, 99.00% and 92.86% on the THDD, Cat Database, Stanford Dogs Dataset and Oxford-IIIT Pet Dataset respectively.
... Face recognition is another example of a biometric modality with a vast body of literature. Older methods, subdivide the pre-processing pipeline of face authentication systems in four steps: face detection [95,101,97], pose estimation [50], landmark detection and finally face alignment based on these landmarks. Face alignment procedures are typically categorized either as template fitting [104] or regression based [12]. ...
... 1) Different from previous face detection surveys [28], [29], [30], [31], [32] in which the content is mainly built on reviewing traditional methods, our survey focuses on deep learning-based face detectors. We have noted the existence of surveys [33], [34], [35] on deep learning; however, they focus on generic object detection, not specifically for face detection. ...
Preprint
Full-text available
Face detection is to search all the possible regions for faces in images and locate the faces if there are any. Many applications including face recognition, facial expression recognition, face tracking and head-pose estimation assume that both the location and the size of faces are known in the image. In recent decades, researchers have created many typical and efficient face detectors from the Viola-Jones face detector to current CNN-based ones. However, with the tremendous increase in images and videos with variations in face scale, appearance, expression, occlusion and pose, traditional face detectors are challenged to detect various "in the wild" faces. The emergence of deep learning techniques brought remarkable breakthroughs to face detection along with the price of a considerable increase in computation. This paper introduces representative deep learning-based methods and presents a deep and thorough analysis in terms of accuracy and efficiency. We further compare and discuss the popular and challenging datasets and their evaluation metrics. A comprehensive comparison of several successful deep learning-based face detectors is conducted to uncover their efficiency using two metrics: FLOPs and latency. The paper can guide to choose appropriate face detectors for different applications and also to develop more efficient and accurate detectors.
... Methods for face recognition and verification (FRV) are becoming increasingly important in many application areas such as healthcare, the gaming industry, user studies, and homeland defense [40,42]. Such methods work reasonably well on face images with standardized (frontal) face positions, good spatial resolution, and controlled lighting conditions, such as found in many imaging benchmarks. ...
Article
Full-text available
Methods for face verification works reasonably well on face images with standardized (frontal) face positions and good spatial resolution. However such methods have significant challenges on poor resolution images, poor lighting conditions and not standard (frontal) face positions. In this paper, we survey the capability of existing face restoration and verification methods, with the aim of understanding how useful face restoration methods are for face verification. We propose a qualitative and quantitative comparison benchmark, and apply it on eight methods for face restoration and six methods for face verification, on several real-world low-quality images from a surveillance context, and outline observed advantages and limitations. Experiments shows that each restoration method can affect each face verification method differently, with fewer than the half of face restoration methods helping face verification. Interestingly, some face restoration methods with less good qualitative evaluation helped face verification the most. Experiments also show that face verification works less good if the resolution decreases.
... It is the local invariant feature for video, and it can resist the changes such as rotation, scale variations, viewpoint change, etc. In the ADMT model, face tracking is done by Kanade-Lucas-Tomasi (KLT) tracker [50], and then Spatio-temporal points from the face can be tracked [51]. The framework is elaborated below subsections. ...
Article
Full-text available
Assistive driving is a complex engineering problem and is influenced by several factors such as the sporadic nature of the quality of the environment, the response of the driver, and the standard of the roads on which the vehicle is being driven. The authors track the driver’s anticipation based on his head movements using Spatio-Temporal Interest Point (STIP) extraction and enhance the anticipation of action accuracy well before using the RNN-LSTM framework. This research tries to tackle a fundamental problem of lane change assistance by developing a novel model called Advanced Driver’s Movement Tracking (ADMT). ADMT uses customized convolution-based deep learning networks by using Recurrent Convolutional Neural Network (RCNN). STIP with eye gaze extraction and RCNN performed in ADMT on brain4cars dataset for driver movement tracking. Its performance is compared with the traditional machine learning and deep learning models, namely Support Vector Machines (SVM), Hidden Markov Model (HMM), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM) and provided an increment of almost 12% in the prediction accuracy and 44% in the anticipation time. ADMT systems outperformed all of the models in terms of both the accuracy of the system and the previously mentioned time of anticipation that is discussed at length in the paper. Thus it assists the driver with additional anticipation time to access the typical reaction time for better preparedness to respond to undesired future behavior. The driver is then assured of a safe and assisted driving experience with the proposed system.
... At such levels, the approach is reading each one of the input images with a window that looks for features in human faces. In the case when collecting enough features, such image's window type is indicated to be a face [7]. For the purpose of obtaining faces of different sizes, then the window should be scaled, while the process will be repeated a lot of times. ...
Article
Full-text available
Automatic face recognition system is suggested in this work on the basis of appearance based features focusing on the whole image as well as local based features focusing on critical face points like eyes, mouth, and nose for generating further details. Face detection is the major phase in face recognition systems, certain method for face detection (Viola-Jones) has the ability to process images efficiently and achieve high rates of detection in real time systems. Dimension reduction and feature extraction approaches are going to be utilized on the cropped image caused by detection. One of the simple, yet effective ways for extracting image features is the Local Binary Pattern Histogram (LBPH), while the technique of Principal Component Analysis (PCA) was majorly utilized in pattern recognition. Also, the technique of Linear Discriminant Analysis (LDA) utilized for overcoming PCA limitations was efficiently used in face recognition. Furthermore, classification is going to be utilized following the feature extraction. The utilized machine learning algorithms are PART and J48. The suggested system is showing high accuracy for detection with Viola-Jones 98.75, whereas the features which are extracted by means of LDA with J48 provided the best results of (F-measure, Recall, and Precision).
... Animal face detection can be used for many security applications that identify animal faces, gender/age detection and visual monitoring [1]. Regretfully, it is still too difficult to detect animal faces given that face textures and shapes are grossly diverse. ...
Preprint
This paper focuses on the face detection problem of three popular animal cat-egories that need control such as horses, cats and dogs. To be precise, a new Convolutional Neural Network for Animal Face Detection (CNNAFD) is actu-ally investigated using processed filters based on gradient features and applied with a new way. A new convolutional layer is proposed through a sparse feature selection method known as Automated Negotiation-based Online Feature Selection (ANOFS). CNNAFD ends by stacked fully connected layers which represent a strong classifier. The fusion of CNNAFD and MobileNetV2 constructs the newnetwork CNNAFD-MobileNetV2 which improves the classification results and gives better detection decisions. Our work also introduces a new Tunisian Horse Detection Database (THDD). The proposed detector with the new CNNAFD-MobileNetV2 network achieved an average precision equal to 99.78%, 99% and 98.28% for cats, dogs and horses respectively.
... The seminal work by Viola and Jones [7] introduces integral image to compute HAAR-like features. For a survey of face detection using hand crafted features, please refer to [8], [9]. ...
Preprint
Full-text available
Tremendous progress has been made on face detection in recent years using convolutional neural networks. While many face detectors use designs designated for the detection of face, we treat face detection as a general object detection task. We implement a face detector based on YOLOv5 object detector and call it YOLO5Face. We add a five-point landmark regression head into it and use the Wing loss function. We design detectors with different model sizes, from a large model to achieve the best performance, to a super small model for real-time detection on an embedded or mobile device. Experiment results on the WiderFace dataset show that our face detectors can achieve state-of-the-art performance in almost all the Easy, Medium, and Hard subsets, exceeding the more complex designated face detectors. The code is available at \url{https://www.github.com/deepcam-cn/yolov5-face}.