The design of intelligent crowd attention detection system based on face detection technology
3
Citation
7
Reference
10
Related Paper
Citation Trend
Abstract:
As a key technology in modern society, face detection has become an active subject in the field of pattern recognition and computer vision. In this paper, we propose some applications of analyzing crowd attention based on face detection technology and realize a crowd attention detection system using Haar-like features and Adaboost algorithm. The system consists of image acquisition module, face detection module, display module and data analysis module. At the same time, we adopt the method of empirical research. Using the camera, combined with software named AMcap, the system captures images of crowed during the display of a video regularly. Then, the system detects the faces which are being concerned about this attractive video in images, circling and counting them. In addition, the system displays the data change graph of the crowd face at different moments, and finally calculates the mathematical expected value of crowd attention for further processing such as Big Data technology. After further processing, the system can obtain the data which reflects the degree of concern in a particular sector. Through the above experiment, Facts have proved that the system can easily, accurately and intuitively detect the crowd attention data and has great value in many sectors.Keywords:
Object-class detection
Human face detection is concerned with finding location and size of every human face in a given image. Locating skin pixels in images or video sequences where people appear has many applications, especially those related to Human-Computer Interaction. Face detection plays a very important role in human computer interaction field. It represents the first step in a fully automatic face recognition, facial features detection, and expression recognition. There are many techniques used in face detection, each one has its advantages and disadvantages. An algorithm is proposed to improve the performance of face detection and eye detection using skin color model under poor illumination condition. Skin regions are extracted using a set of bounding rules based on the skin color distribution obtained from a training set. After detecting the face region, the Hough transform is used in order to detect the facial feature i. e. eye, specially circular Hough transform is used in order to detect the circular object in the image.
Feature (linguistics)
Cite
Citations (3)
A novel human face tracking system with multiple cameras is proposed in this article. Unlike most surveillance approaches requiring overlapping field of views (FOVs) for tracking targets across multiple cameras, our proposed system utilizes face recognition as cue to build correspondence between cameras, and is therefore compatible with application with non-overlapping FOVs. Face recognition achieves great success in recent years, however it has an intrinsic limitation of low tolerance to face pose changes. In the proposed system, face reconstruction technology is adopted to transform non-frontal human face to a frontal one to substantially enhance the usability of face recognition, so that correspondence of humans across multiple trajectories can be reliably established.
Facial motion capture
Tracking (education)
Three-dimensional face recognition
Object-class detection
Cite
Citations (0)
In this paper, we present a real-time face pose recognition system using a stereo camera. The face pose recognition is to measure the direction of a face by extracting the face region in the input image. In this paper, we detect an apparent foreground object region in a certain distance using a stereo camera. We then remove background region, and extract the object’s face region using color information. Finally, the face pose can be recognized by analyzing the face distribution ratio. Unlike various existing pose recognition methods, the proposed technique does not require any training algorithms, which makes real-time applications possible.
Three-dimensional face recognition
3D single-object recognition
Object-class detection
Cite
Citations (4)
This paper presents a real-time face recognition system. The system uses a stereo camera to locate, track, and recognize a person's face. Our algorithm improves state-of-the-art monocular 2D object recognition techniques by additionally considering the facial 3D surface, which is relatively stable under different lighting conditions. First, faces are detected and their surfaces are reconstructed from the stereo images. Afterwards, a 3D face is composed by joining 2D image data and appropriate depth data. The 3D face is then decomposed into its principal components. The principal components are used to recognize a 3D face by comparing characteristics of the current face to those of known individuals in a database. The result is an efficient and accurate face recognition algorithm. To evaluate our approach, we compared its performance to a classical monocular face recognition algorithm and observed that the recognition rate increased on average by 7.7 percent.
Three-dimensional face recognition
Monocular
3D single-object recognition
Object-class detection
Face hallucination
Cite
Citations (7)
Face profile is an important aspect of face recognition and it provides a complementary structure of the face that is seen in the non-frontal view. In the past, several methods have been proposed to recognize face profiles in still images. However, face profile images that are captured at a distance by surveillance cameras usually are video sequences that have a low resolution. It is difficult to extract accurate face profile directly from a low-resolution video frame, which does not have many pixels on the face profile. The emphasis of this paper is to introduce a practical approach for human recognition by using high-resolution face profile images constructed from the low-resolution videos. We use both the spatial and temporal information present in a number of adjacent low-resolution frames of a video sequence to construct high-resolution face profile images. As the quality of high-resolution images relies on the correctness of image alignment between consecutive frames, an elastic registration algorithm is used for face profile image alignment. A match statistic is designed to detect and discard poorly aligned images which may degrade the quality of the high-resolution face profile image. After obtaining high-resolution face profile images, we use a dynamic time warping method for face profile recognition. A number of dynamic video sequences are tested to demonstrate the applicability and reliability of our method.
Image warping
Object-class detection
Three-dimensional face recognition
Cite
Citations (14)
Human face detection plays an important role in applications such as video surveillance, human computer interface, face recognition, and face image database management, etc. We propose a face detection algorithm for color images in complex backgrounds, using fuzzy logic, fast marching method and some image processing techniques. The algorithm is mainly based on skin colors. Sizes, shapes of faces and facial features are fuzzy factors for verifying face candidates. To reduce computation time, we deal only with border points of each object. Regardless of positions of true faces detected, we also get their relatively accurate borders for the results. Experimental results demonstrate successful face detection over a wide range of facial variations in color, position, scale, orientation, 3D pose, and expression in images from several photo collections (both indoors and outdoors). The computation time is also faster than many other face algorithms based on skin colors.
Object-class detection
Three-dimensional face recognition
Position (finance)
Cite
Citations (11)
This paper proposes a recognition and face tracking based on computer vision techniques using OpenCV libraries by applying multiple phases in cascade. The algorithm allows for a more robust tracking because it combines face and eye detection. Besides, it detects edges and cuts the region of interest (ROI) where the face is located. After that, the algorithm verifies if there is any face in the ROI passing again the face and eye detector. The identification is done through a comparison of the detected face and a stored database of images.
Three-dimensional face recognition
Object-class detection
Facial motion capture
Identification
Tracking (education)
Region of interest
Face hallucination
Cite
Citations (0)
Face detection is the essential front end of any face recognition system, which locates and segregates face regions from cluttered images, either obtained from video or still image. It also has numerous applications in areas like surveillance and security control systems, content-based image retrieval, video conferencing and intelligent human-computer inter-faces. Most of the current face recognition systems presume that faces are readily available for processing. However, in reality, we do not get images with just faces. We need a system, which will detect, locate and segregate faces in cluttered images, so that these segregated faces can be given as input to face recognition systems. In this paper, we address the problem of face detection in still images by using skin color region extraction and adaptive template matching method.
Object-class detection
Template matching
Three-dimensional face recognition
Cite
Citations (2)
Images taken of human faces vary in perspective depending on the view point of the observer. Few of the face recognition methods reported in the literature are capable of recognising faces under varying poses. To improve the ability of face recognition systems to recognise faces under varying poses, a novel method is proposed, that takes a face image of arbitrary pose and synthesises a standard frontal pose image of the same face.
Three-dimensional face recognition
Observer (physics)
Object-class detection
Cite
Citations (2)
This paper provides a method for detection and recognition face of man using 3D face extracting and applying 3D features.In this paper, it is used from the Bio-ID package contained 1520 images of man face.A number of these images were used for 2D face modeling; another number was used for 3D face modeling, and another number was used for test.At first, the landmarks on the face image are determined, semi automatically.Then the shape, texture and appearance model of face images is constructed.Using these models and the fast Active Appearance Model search, the landmarks on the test image are determined.More ever, from 24 3D images, obtained by a 3D scanner, the variations of shape, texture and appearance are modeled.Using the 3D models, 2D landmarks and an 3D Initialized Active Appearance Model Search method (3D IAAMS), the 3D frame of face (this frame is described by 3D landmarks) in an image is constructed.These 3D frames with the texture are used for face recognition.
Cite
Citations (0)