• Journal of Electrical Engineering and Technology
• /
• v.13 no.1
• /
• pp.485-492
• /
• 2018

In this paper, we propose facial expression recognition using CNN (Convolutional Neural Network), one of the deep learning technologies. The proposed structure has general classification performance for any environment or subject. For this purpose, we collect a variety of databases and organize the database into six expression classes such as 'expressionless', 'happy', 'sad', 'angry', 'surprised' and 'disgusted'. Pre-processing and data augmentation techniques are applied to improve training efficiency and classification performance. In the existing CNN structure, the optimal structure that best expresses the features of six facial expressions is found by adjusting the number of feature maps of the convolutional layer and the number of nodes of fully-connected layer. The experimental results show good classification performance compared to the state-of-the-arts in experiments of the cross validation and the cross database. Also, compared to other conventional models, it is confirmed that the proposed structure is superior in classification performance with less execution time.

• Journal of Broadcast Engineering
• /
• v.22 no.2
• /
• pp.162-172
• /
• 2017

In this paper, we propose facial expression recognition using CNN (Convolutional Neural Network), one of the deep learning technologies. To overcome the disadvantages of existing facial expression databases, various databases are used. In the proposed technique, we construct six facial expression data sets such as 'expressionless', 'happiness', 'sadness', 'angry', 'surprise', and 'disgust'. Pre-processing and data augmentation techniques are also applied to improve efficient learning and classification performance. In the existing CNN structure, the optimal CNN structure that best expresses the features of six facial expressions is found by adjusting the number of feature maps of the convolutional layer and the number of fully-connected layer nodes. Experimental results show that the proposed scheme achieves the highest classification performance of 96.88% while it takes the least time to pass through the CNN structure compared to other models.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.944-947
• /
• 2022

Due to the coronavirus pandemic, the wearing of a mask has been increasing worldwide; thus, the importance of image analysis on masked face images has become essential. Although head pose estimation can be applied to various face-related applications including driver attention, face frontalization, and gaze detection, few studies have been conducted to address the performance degradation caused by masked faces. This study proposes a new data augmentation that synthesizes the masked face, depending on the face image size and poses, which shows robust performance on BIWI benchmark dataset regardless of mask-wearing. Since the proposed scheme is not limited to the specific model, it can be utilized in various head pose estimation models.

• Journal of Sasang Constitutional Medicine
• /
• v.34 no.3
• /
• pp.31-40
• /
• 2022

Objectives Sasang constitutional medicine is a traditional Korean medicine that classifies humans into four constitutions in consideration of individual differences in physical, psychological, and physiological characteristics. In this paper, we proposed a method to classify Taeeum person (TE) and Non-Taeeum person (NTE), Soeum person (SE) and Non-Soeum person (NSE), and Soyang person (ST) and Non-Soyang person (NSY) using a convolutional neural network with only facial images. Methods Based on the convolutional neural network VGG16 architecture, transfer learning is carried out on the facial images of 3738 subjects to classify TE and NTE, SE and NSE, and SY and NSY. Data augmentation techniques are used to increase classification performance. Results The classification performance of TE and NTE, SE and NSE, and SY and NSY was 77.24%, 85.17%, and 80.18% by F1 score and 80.02%, 85.96%, and 72.76% by Precision-Recall AUC (Area Under the receiver operating characteristic Curve) respectively. Conclusions It was found that Soeum person had the most heterogeneous facial features as it had the best classification performance compared to the rest of the constitution, followed by Taeeum person and Soyang person. The experimental results showed that there is a possibility to classify constitutions only with facial images. The performance is expected to increase with additional data such as BMI or personality questionnaire.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.51-59
• /
• 2024

Collecting a dataset with a corresponding labeled gaze vector requires a high cost in the gaze estimation field. In this paper, we suggest a data augmentation of manipulating the gaze of an original image, which improves the accuracy of the gaze estimation model when the number of given gaze labels is restricted. By conducting multi-class gaze bin classification as an auxiliary task and adjusting the latent variable of the diffusion model, the model semantically edits the gaze from the original image. We manipulate a non-binary attribute, pitch and yaw of gaze vector to a desired range and uses the edited image as an augmented train data. The improved gaze accuracy of the gaze estimation network in the semi-supervised learning validates the effectiveness of our data augmentation, especially when the number of gaze labels is 50k or less.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.47 no.4
• /
• pp.286-290
• /
• 2021

Objectives: The use of fat grafts in maxillofacial sculpturing is currently a common technique. Unlike fillers, autologous fats unite with facial tissues, but long-term results may still be unsatisfactory. Sharing long-term follow-ups can be helpful in making outcomes more predictable. Materials and Methods: The data from patients who were admitted from 2014 to 2016 for fat augmentation were collected. In all cases, fat grafts were injected by blunt cannula using a tunneling technique in different planes. A fan shape order for the malar, periorbital, nasolabial fold, mandibular angle and body, and perioral area was established. Results: Autologous fat was used for different sites of the maxillofacial regions. Of 15 patients, two patients were not satisfied due to fat graft resorption. For this, further injections were performed six months after the first injection using preserved fat grafts. One patient continued to be dissatisfied. There were no other complications related to fat transplants. Conclusion: Fat transplantation is a safe, reliable, and non-invasive method for facial contour and facial soft tissue defect restoration. Additional methods such as mesenchymal stem cells along with fat injection increase the survival rate of transferred fat.

• Archives of Craniofacial Surgery
• /
• v.11 no.2
• /
• pp.77-84
• /
• 2010

Purpose: The nasal bones are the most common fracture sites of the facial bones, and a careful reduction may still result in secondary deformities, such as saddle nose, deviated nose, hump nose etc, requiring secondary cosmetic rhinoplasty. Therefore, this study examined the clinical characteristics of nasal bone fractures to propose guidelines for patient selection and surgical procedures to achieve more satisfactory results and to prevent secondary deformities with simultaneous augmentation rhinoplasty and bony reduction. Methods: The study was based on 26 out of 149 nasal bone fracture patients who underwent simultaneous augmentation rhinoplasty with bony reduction between May 2008 and April 2009. Retrospective analysis was performed according to the clinical data, surgical techniques and postoperative results. Results: Of the 26 patients, there were 15 males and 11 females. The incidence according to the Stranc's classification revealed that 62% of patients were injured by a frontal impact and 38% by a lateral impact. Frontal impact plane I (50%) was the most frequent type. At the follow up, 18 (81.2%) out of 22 patients were satisfied with their postoperative outcome, and the remaining 4 patients were fair. No one was dissatisfied. However, 5 cases in 3 patients (23%) had some complications; minimal implant deviation in 2 cases, minor irregularity on the nasal dorsum in 2 cases and palpable implant movement under palpation in 1 case. None of these cases required surgical correction. Conclusion: With the proper guidance, simultaneous augmentation rhinoplasty with bony reduction can prevent secondary deformities and satisfy the cosmetic outcomes.

• Archives of Plastic Surgery
• /
• v.35 no.5
• /
• pp.597-602
• /
• 2008

Purpose: Presently, silicone rubber is chosen most frequently for nasal augmentation. However, there is a possibility of extrusion with this material. Sometimes, noses are prone to be traumatized, and then silicone rubber has a possibility of deformity or deviation resulting in trauma. We experienced cases with complications and traumatic deformities after the augmentation rhinoplasty. Methods: A retrospective review was performed to determine the characteristics of the implanted nasal silicone prosthesis after trauma. The patients' data such as deviation of implant, shape of fracture, age and sex of the patient, time of treatment, operative methods were reviewed. From March 2001 to March 2008, this study was performed in 30 patients. The patients were 25 females and 5 males, from 24 to 60 years of age, with an average of 42. All patients had previous augmentation rhinoplasty with silicone implant. Results: All of the 30 patients were confirmed as deviation of silicone and nasal bone fractures in the facial bone CT scan. The most common cause of fracture was traffic accident. The classification of nasal trauma after augmentation was done by facial bone CT. Class I: Deviation of silicone without nasal bone fracture without extrusion(12 cases, 40%), Class II: Deviation of silicone without nasal bone fracture and with extrusion(4 cases, 13%), Class III: Deviation of silicone with nasal bone fracture and without extrusion(8 cases, 27%), Class IV: Deviation of silicone with nasal bone fracture and with extrusion(3 cases, 10%), Class V: Mild deviation of silicone with nasal bone fracture(3cases, 3%). Specially, the comminuted or trapezoid nasal fracture was confirmed in 11 cases(Class III, IV). Conclusion: The problems of silicone implant have generally been related to foreign body reactions, rigidity of the material, encapsulation, infections, and extrusion. We experienced 11 cases of comminuted or comminuted trapezoid shaped fracture below nasal implant. So, we think this phenomenon could be used in late problem of silicone implant.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.17-18
• /
• 2016

본 논문에서는 딥러닝 기술 중의 하나인 CNN(Convolutional Neural Network) 기반의 얼굴 표정 인식 기법을 제안한다. 제안한 기법에서는 획득한 여섯 가지 주요 표정의 얼굴영상들을 학습 데이터로 이용할 때 분류 성능을 저해시키는 과적합(over-fitting) 문제를 해결하기 위해서 데이터 증대 기법(data augmentation)을 적용한다. 또한 기존의 CNN 구조에서 convolutional layer 및 node의 수를 변경하여 학습 파라미터 수를 대폭 감소시킬 수 있다. 실험 결과 제안하는 데이터 증대 기법 및 개선한 구조가 높은 얼굴 표정 분류 성능을 보여준다는 것을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.33-36
• /
• 2021

본 논문에서는 인간의 표정과 목소리를 기반으로 한 감정 분석기를 제안한다. 제안하는 분석기들은 수많은 인간의 표정 중 뚜렷한 특징을 가진 표정 7가지를 별도의 클래스로 구성하며, DNN 모델을 수정하여 사용하였다. 또한, 음성 데이터는 학습 데이터 증식을 위한 Data Augmentation을 하였으며, 학습 도중 과적합을 방지하기 위해 콜백 함수를 사용하여 가장 최적의 성능에 도달했을 때, Early-stop 되도록 설정했다. 제안하는 표정 감정 분석 모델의 학습 결과는 val loss값이 0.94, val accuracy 값은 0.66이고, 음성 감정 분석 모델의 학습 결과는 val loss 결과값이 0.89, val accuracy 값은 0.65로, OpenCV 라이브러리를 사용한 모델 테스트는 안정적인 결과를 도출하였다.