• 융합신호처리학회논문지
• /
• 제2권4호
• /
• pp.16-21
• /
• 2001

본 논문에서는 2매의 2차원 얼굴영상으로부터 이들의 합성하여 3차원 얼굴의 가상형상을 복원한다. 여기서 2매의 2차원 얼굴영상은 정면과 측면 영상을 사용한다. 임의의 일반 얼굴에 대한 기준모델을 정하고 이 모델에서 얼굴형상의 특징을 표현하는 귀 , 눈 코 및 입 부분에 집중적으로 특징 점을 규정한다. 그 이외에 이마 및 턱부분에도 특징 점을 지정하여 그 위치 좌표를 저장해 둔다. 그 후 정면 영상의 좌 우측에 측면영상을 대칭적으로 접속하고 영상의 기하 변환방법을 적용하여 점차적으로 합성한다. 합성을 효과적으로 처리하기 위해 측면 영상을 정면 영상에 정합될 수 있도록 기하변환을 이용한다. 이 때 나타나는 합성부분에 색상 및 명도의 차를 제거하기 위해 스므딩필터(3$\times$3 화소의 마스크)를 적용하여 자연스런 3차원 가상얼굴을 구현하게 된다. 그 결과 불특정 얼굴형상도 3차원으로 구현할수 있음을 확인하였다.

• 한국정보통신학회논문지
• /
• 제7권2호
• /
• pp.328-333
• /
• 2003

본 논문에서는 얼굴의 표정 변화에 영향을 주는 해부학에 기반한 18개의 근육군쌍을 바탕으로 하여 얼굴 표정 애니메이션을 위한 근육의 움직임을 조합할 수 있도록 하였다. 개인의 이미지에 맞춰 메쉬를 변형하여 표준 모델을 만든 다음, 사실감을 높이기 위해 개인 얼굴의 정면과 측면 2장의 이미지를 이용하여 메쉬에 매핑하였다. 얼굴의 표정 생성을 애니메이션 할 수 있는 원동력이 되는 근육 모델은 Waters의 근육모델을 수정하여 사용하였다. 이러한 방법을 사용하여 텍스처가 입혀진 변형된 얼굴을 생성하였다. 또한, Ekman이 제안한 6가지 얼굴 표정을 애니메이션 하였다.

• 경영과학
• /
• 제33권3호
• /
• pp.65-87
• /
• 2016

Being the FinTech technologies rapidly developed, the non face-to-face private loan market is also growing dramatically. While the real-world interests in this market are keen, the empirical studies on the issue are few compared to its prospective impact on credit loan market. This paper suggests a credit scoring model for the non face-to-face private loan employing the ratings approach (the absolute measurement method) of AHP. Analyzing a sample of data consisting of 460,000 transaction records over an 8-year period in the United States, we develop a scoring model for the non face-to-face private loan screening, and validate the model for the practical usage. Conducting sensitivity analysis, we suggest customized cut-off points for the loan execution to suit each individual loan institution's need.

• 한국정보전자통신기술학회논문지
• /
• 제9권6호
• /
• pp.546-552
• /
• 2016

We present a quantitative evaluation of an algorithm for model-based face recognition. The algorithm actively learns how individual faces vary through video sequences, providing on-line suppression of confounding factors such as expression, lighting and pose. By actively decoupling sources of image variation, the algorithm provides a framework in which identity evidence can be integrated over a sequence. We demonstrate that face recognition can be considerably improved by the analysis of video sequences. The method presented is widely applicable in many multi-class interpretation problems.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.493-494
• /
• 2022

The development of livestock faces intensive farming results in a rising need for recognition of individual animals such as cows and pigs is related to high traceability. In this paper, we present a non-invasive biometrics systematic approach based on the deep-learning classification model to pig face identification. Firstly, in our systematic method, we build a ROS data collection system block to collect 10 pig face data images. Secondly, we proposed a preprocessing block in that we utilize the SSIM method to filter some images of collected images that have high similarity. Thirdly, we employ the improved image classification model of CNN (ViT), which uses the finetuning and pretraining technique to recognize the individual pig face. Finally, our proposed method achieves the accuracy about 98.66%.

• 한국정보통신학회논문지
• /
• 제7권2호
• /
• pp.322-327
• /
• 2003

본 논문에서는 효율적인 얼굴 애니메이션을 구축하기 위하여 동적인 근육 모델을 기반으로 한 3차원 얼굴 모델링 기법을 제안하였다. 동적인 근육 모델을 기반으로 얼굴 근육을 256개의 점과 이 점들을 연결한 faceline으로 구성한 wireframe을 구축하고, wireframe을 이용한 표준 모델을 구성한 후 정면과 측면의 두장의 2D 영상을 사용하여 텍스처 매핑을 실시하여 3차원 개인 얼굴 모델을 생성하였다. 정확한 매핑을 위하여 특징점들의 정면과 측면 부분을 이용했는데, 정면 이미지와 정면 특징점들의 2차원 좌표를 이용하여 텍스터 좌표를 가진 얼굴을 만든 다음, 측면 이미지와 측면 특징점들의 2차원 좌표를 이용하여 텍스터 좌표를 가진 얼굴을 구축하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -2
• /
• pp.814-817
• /
• 2002

High-dimensional data with two or more attributes are considered. A typical example of such data is face images of various individuals and expressions. In these cases, collecting a complete data set is often difficult since the number of combinations can be large. In the present study, we propose a method to interpolate data of missing combinations from other data. If this becomes possible, robust recognition of multiple attributes is expectable. The key of this subject is appropriate extraction of the similarity that the face images of same individual or same expression have. Bilinear model ［1］has been proposed as a solution of this subjcet. However, experiments on application of bilinear model to classification of face images resulted in low performance ［2］. In order to overcome the limit of bilinear model, in this research, a nonlinear model on a neural network is adopted and usefulness of this model is experimentally confirmed.

• 한국의료질향상학회지
• /
• 제30권1호
• /
• pp.55-75
• /
• 2024

Purpose: This study aimed to elucidate the intention to use non-face-to-face treatment, which was temporarily allowed during the coronavirus disease 2019 pandemic. Based on the social cognitive theory, individual behavioral changes occur through the dynamic interaction of individual, environmental, and behavioral factors. Thus, we investigated the impact of personal, environmental, and behavioral factors on the acceptance of non-face-to-face treatment. Methods: A Web survey was conducted using Korea Research Panel between December 26 and 29, 2022, to examine the conceptual framework. The survey targeted adults aged 19 and older, regardless of whether they had used non-face-to-face treatment. A total of 502 responses were collected. Further, a three-step hierarchical regression analysis was conducted using SPSS Windows software version 25.0. Results: The study showed that 131 out of 502 respondents had experience using non-face-to-face treatment, while 371 did not. The factors that influenced the intention to accept non-face-to-face treatment included the general characteristics of the participants (women, underlying disease), personal factors (usefulness, cost savings, knowledge), and environmental factors (social norms, trust, perceived risk). The model demonstrated an explanatory power of 65%. Conclusion: The results of this study directly show that intention is linked to behavior through the interaction between personal and environmental factors. Further research is needed to explore additional factors influencing the intention to accept non-face-to-face treatment, enabling its effective use in preventing and treating various diseases, including infectious diseases.

• Archives of Plastic Surgery
• /
• 제37권4호
• /
• pp.361-368
• /
• 2010

Purpose: Authors tried to analyze the influence of individual facial aesthetic subunits on the cognition of facial attractiveness in public and suggest a mathematical model which explain the facial attractiveness. Methods: Independent facial aesthetic subunits are extracted from facial photographs from three women (11 frontal and 7 lateral aesthetic subunits). Each facial subunits of three women are rated in terms of relative rank by 164 peoples (68 man and 96 woman, average age was 32.4, and ranged ${\pm}$ 9.8 years). $x^2$-test and categorical regression analysis were performed. Results: There was no difference in the aesthetic preference in terms of ages or sexes in large. Beautification of individual aesthetic subunits can predict the overall facial attractiveness up to 42.1% in frontal face (Adjusted $R^2$=0.421, F=6.39, p=0.000 < 0.05) and 22.7% in lateral face (Adjusted $R^2$=0.227, F=4.42, p=0.000 < 0.05). Aesthetic appearance of eyes (p=0.001), upper face (p=0.034) in frontal face and midface (p=0.000) in lateral face are statistically important factors in the cognition of facial attractiveness. Conclusion: Authors experimently proved that harmony and balance among facial aesthetic subunits are the most important factors, in embarking on facial aesthetic plastic surgery, for better enhancement of facial attractiveness.

• 한국통신학회논문지
• /
• 제30권9C호
• /
• pp.887-898
• /
• 2005

본 논문에서는 피로를 인식하기 위하여 베이지안 네트워크를 기반으로 한 확률 모델을 제안하고자 한다. 먼저 적외선 조명을 조사하여 눈거풀의 움직임, 시선 방향, 얼굴의 움직임 및 얼굴 표정 같은 얼굴특징정보를 측정하였다. 그러나 각각의 얼굴특징정보만으로 생체 피로를 결정하기에는 충분하지 않다. 그러므로, 본 논문에서는 생체 피로를 확률적 추론하기 위하여 가능한 많은 피로 원인에 대한 정보와 얼굴특징정보들로 베이지안 네트워크 모델을 구성하여 BN 피로지수를 산출하였다. 또한, BN 피로지수의 문턱치값은 MSBNX 시물레이션 결과 0.95로 산출되었다. 실험 결과 BN 피로지수와 TOVA 응답 시간을 비교한 결과 밀접한 상관관계가 있음을 확인하여 제안한 피로인식모델의 유효성을 입증하였다.