• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1020-1025
• /
• 2021

Facial alignment is very important task for human life. And facial landmark detection is one of the instrumental methods in face alignment. We introduce the stacked hourglass networks with transposed convolutional layers for facial landmark detection. our method substitutes nearest neighbor upsampling for transposed convolutional layer. Our method returns better accuracy in facial landmark detection compared to stacked hourglass networks with nearest neighbor upsampling.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3685-3700
• /
• 2015

To avoid influences caused by pose, illumination and facial expression variations, we propose a robust three-step algorithm based on ASM and POEM for facial landmark localization. Firstly, Model Selection Factor is utilized to achieve a pose-free initialized shape. Then, we use the global shape model of ASM to describe the whole face and the texture model POEM to adjust the position of each landmark. Thirdly, a second localization is presented to discriminatively refine the subtle shape variation for some organs and contours. Experiments are conducted in four main face datasets, and the results demonstrate that the proposed method accurately localizes facial landmarks and outperforms other state-of-the-art methods.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.666-673
• /
• 2014

This paper proposes a scalable face recognition method for unconstrained face databases, and shows a simple experimental result. Existing face recognition research usually has focused on improving the recognition rate in a constrained environment where illumination, face alignment, facial expression, and background is controlled. Therefore, it cannot be applied in unconstrained face databases. The proposed system is face feature extraction algorithm for unconstrained face recognition. First of all, we extract the area that represent the important features(landmarks) in the face, like the eyes, nose, and mouth. Each landmark is represented by a high-dimensional LBP(Local Binary Pattern) histogram feature vector. The multi-scale LBP histogram vector corresponding to a single landmark, becomes a low-dimensional face feature vector through the feature reduction process, PCA(Principal Component Analysis) and LDA(Linear Discriminant Analysis). We use the Rank acquisition method and Precision at k(p@k) performance verification method for verifying the face recognition performance of the low-dimensional face feature by the proposed algorithm. To generate the experimental results of face recognition we used the FERET, LFW and PubFig83 database. The face recognition system using the proposed algorithm showed a better classification performance over the existing methods.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.539-554
• /
• 2018

Effective interaction between user and device is considered an important ability of IoT devices. For some applications, it is necessary to recognize human facial expressions in real time and make accurate judgments in order to respond to situations correctly. Therefore, many researches on facial image analysis have been preceded in order to construct a more accurate and faster recognition system. In this study, we constructed an automatic recognition system for facial expressions through two steps - a facial recognition step and a classification step. We compared various models with different sets of data with pixel information, landmark coordinates, Euclidean distances among landmark points, and arctangent angles. We found a fast and efficient prediction model with only 30 principal components of face landmark information. We applied several prediction models, that included linear discriminant analysis (LDA), random forests, support vector machine (SVM), and bagging; consequently, an SVM model gives the best result. The LDA model gives the second best prediction accuracy but it can fit and predict data faster than SVM and other methods. Finally, we compared our method to Microsoft Azure Emotion API and Convolution Neural Network (CNN). Our method gives a very competitive result.

• Journal of information and communication convergence engineering
• /
• v.11 no.3
• /
• pp.207-215
• /
• 2013

In this paper, we address the challenging computer vision problem of obtaining a reliable facial expression analysis from a naturally interacting person. We propose a system that combines a 3D generic face model, 3D head tracking, and 2D tracker to track facial landmarks and recognize expressions. First, we extract facial landmarks from a neutral frontal face, and then we deform a 3D generic face to fit the input face. Next, we use our real-time 3D head tracking module to track a person's head in 3D and predict facial landmark positions in 2D using the projection from the updated 3D face model. Finally, we use tracked 2D landmarks to update the 3D landmarks. This integrated tracking loop enables efficient tracking of the non-rigid parts of a face in the presence of large 3D head motion. We conducted experiments for facial expression recognition using both framebased and sequence-based approaches. Our method provides a 75.9% recognition rate in 8 subjects with 7 key expressions. Our approach provides a considerable step forward toward new applications including human-computer interactions, behavioral science, robotics, and game applications.

• Journal of Broadcast Engineering
• /
• v.24 no.4
• /
• pp.564-572
• /
• 2019

With the development of deep learning, the performance of facial landmark detection methods has been greatly improved. The heat map regression method, which is a representative facial landmark detection method, is widely used as an efficient and robust method. However, the landmark coordinates cannot be directly obtained through a single network, and the accuracy is reduced in determining the landmark coordinates from the heat map. To solve these problems, we propose to combine integral regression with the existing heat map regression method. Through experiments using various datasets, we show that the proposed integral regression network significantly improves the performance of facial landmark detection.

• Archives of Plastic Surgery
• /
• v.32 no.3
• /
• pp.311-313
• /
• 2005

Generally, many Asian women tend to dislike the square jaw, as they believe it makes the face look wider, giving a stubborn and strong impression. Contouring of the mandible is therefore a relatively common aesthetic procedure among Asians. These days, the use of botulinum toxin for contouring of the lower face offer simple alternative to surgery. Motor point, which is the site over a muscle where its contraction may be elicited by a minimal intensity short duration electrical stimulus, is the optimal injection point of botulinum toxin. Study was undertaken to identify the location of motor point of the masseter muscle and the skin surface landmark. First, the thickest point of the masseter muscle was inspected through palpation and inspection by 3 different individual plastic surgeons and then compound muscle action potentials(CMAPs) of masseter muscle in 15 health volunteers were recorded using EMG. For the localization of the measured points, line between lateral canthus to the mandibular angle was used. Location of motor points were mapped to skin surface from lateral canthus in a percentage of the distance along the landmark line and in distance in millimeters. The clinical injection point was located at 71.69 percentile and 7.3 mm of the landmark line. The motor point test was located at 72.54 percentile and 7.1 mm of the landmark line. The depth of motor point was 16mm. There was no statistically significant difference between the clinical injection point and the motor point. We conclude that surface mapping of motor point of the masseter muscle would increase accessibility and accuracy in botulinum toxin injection for contouring of the lower face.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.163-170
• /
• 2022

This paper introduces a method of extracting facial features due to unrefined inputs in real life and improving the problem of not guaranteeing the ideal performance and speed of the object recognition model through a storage algorithm through weight summation. Many facial recognition processes ensure accuracy in ideal situations, but the problem of not being able to cope with numerous biases that can occur in real life is drawing attention, which may soon lead to serious problems in the face recognition process closely related to security. This paper presents a method of quickly and accurately recognizing faces in real time by comparing feature points extracted as input with a small number of feature points that are not overfit to multiple biases, using that various variables such as picture composition eventually take an average form.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.239-240
• /
• 2020

마스크 쓴 얼굴에 대해 랜드마크 분석을 진행하기 위해서는 대량의 마스크가 착용된 얼굴 데이터셋이 필요하다. 본 논문에서는 공개된 얼굴 데이터셋에 자동으로 마스크를 합성하여 대량의 마스크를 착용한 얼굴 데이터셋을 생성하는 시스템을 제안한다. 마스크는 얼굴의 많은 부분을 가리는 물체이다. 따라서 마스크를 쓴 얼굴에 대해서는 일반적인 얼굴 데이터셋으로 학습된 landmark detector가 잘 작동하지 않는다. landmark detector가 잘 작동하게 하려면 마스크를 쓴 얼굴에 대해서 학습을 시켜야 한다. 그러나 현재 마스크를 쓴 얼굴 이미지와 풍부한 landmark 정보를 함께 가지고 있는 데이터셋이 존재하지 않기 때문에 학습에 어려움이 있다. 이 문제를 해결하기 위해 마스크 얼굴 이미지 데이터셋을 만들어내는 방법을 제안하고 마스크를 착용한 얼굴에도 잘 작동하는 랜드마크 검출기를 학습시켜 그 효용을 입증하였다.