• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.197-204
• /
• 2008

It is well known that the tracking a certain person is a vary needed technic in the humanoid robot. In robot technic, we should consider three aspects that is cloth color matching, face recognition and motion analysis. Because a robot technic use some sensors, it is many different with the robot technic to track a certain person through the CCTV images. A system speed should be fast in CCTV images, hence we must have small calculation numbers. We need the statistical variable for color matching and we adapt the eigen-face for face recognition to speed up the system. In this situation, motion analysis have to added for the propose of the efficient detecting system. But, in many motion analysis systems, the speed and the recognition rate is low because the system operates on the all image area. In this paper, we use the moving energy only on the face area which is searched when the face recognition is processed, since the moving energy has low calculation numbers. When the proposed algorithm has been compared with Girondel, V. et al's method for experiment, we obtained same recognition rate as Girondel, V., the speed of the proposed algorithm was the more faster. When the LDA has been used, the speed was same and the recognition rate was better than Girondel, V.'s method, consequently the proposed algorithm is more efficient for tracking a certain person.

• Journal of Internet Computing and Services
• /
• v.7 no.3
• /
• pp.143-153
• /
• 2006

Facial feature detection is a basic technology in applications such as human computer interface, face recognition, face tracking and image database management. The speed of feature detection algorithm is one of the main issues for facial feature detection in real-time environment. Primary factors like a variation by lighting effect, location, rotation and complex background give an effect to decrease a detection ratio. A facial feature detection algorithm is proposed to improve the detection ratio and the detection speed. The proposed algorithm detects skin regions over the entire image improved by CLAHE, an algorithm for light compensation against varying lighting conditions. To extract facial feature points on detected skin regions, it uses appearance-based geometrical characteristics of a face. Since the method shows fast detection speed as well as efficient face-detection ratio, it can be applied in real-time application to face tracking and face recognition.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.629-631
• /
• 2016

This paper dealt with the traveling robot for a robot to recognize a face.. This is to be carried out in a robotics basically. At the same time it can be described as utilizing available technology in various fields.In this paper, we envisioned that the operation of the face recognition biometric technology as a security measure in a particular space on the mobile robot.

• Journal of Internet Computing and Services
• /
• v.6 no.4
• /
• pp.93-102
• /
• 2005

In Recent times, Netizens have frequently use virtual character 'Avatar' schemes in order to present their own identity, there is a strong need for avatars to resemble the user. This paper proposes an extraction technique for facial region and features that are used in generating the avatar automatically. For extraction of facial feature component, the method uses ACM and edge information. Also, in the extraction process of facial region, the proposed method reduces the effect of lights and poor image quality on low resolution pictures. this is achieved by using the variation of facial area size which is employed for external energy of ACM. Our experiments show that the success rate of extracting facial regions is $92{\%}$ and accuracy rate of extracting facial feature components is $83.4{\%}$, our results provide good evidence that the suggested method can extract the facial regions and features accurately, moreover this technique can be used in the process of handling features according to the pattern parts of automatic avatar generation system in the near future.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.629-632
• /
• 2012

클로즈 업 숏은 영화의 서사구조와 밀접한 관련이 있다. 등장인물의 얼굴을 확대하여 보여줌으로써 표정의 변화를 섬세하게 표현할 수 있다. 이를 통해 관객에게 등장인물의 감정이나 심리상태를 세밀하게 묘사해 줄 수 있다. 따라서 영화에서 클로즈 업 숏을 검출하고 분포도를 파악하는 것은 영화의 서사구조를 분석하는데 있어 중요한 데이터로 활용될 수 있다. 또한 영화의 정보 검색에 활용될 수도 있다. 본 논문에서는 얼굴 검출 기술을 활용하여 클로즈 업 숏을 자동으로 검출하여 영화의 클로즈 업숏 분포도를 계산하는 방법을 제안한다. 또한 제안한 방법을 시스템으로 개발하여 서사구조와 클로즈업 숏 간의 관계를 분석한다. 10개의 영화에 대한 실험을 통해 클로즈 업 숏이 영화의 전반부보다 후반부로 갈수록 많이 사용되는 것을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.559-561
• /
• 2022

인공지능 시대에 들어서면서 개인 맞춤형 환경을 제공하기 위하여 사람의 감정을 인식하고 교감하는 기술이 많이 발전되고 있다. 사람의 감정을 인식하는 방법으로는 얼굴, 음성, 신체 동작, 생체 신호 등이 있지만 이 중 가장 직관적이면서도 쉽게 접할 수 있는 것은 표정이다. 따라서, 본 논문에서는 정확도 높은 얼굴 감정 식별을 위해서 Convolution Block Attention Module(CBAM)의 각 Gate와 Residual Block, Skip Connection을 이용한 Identity- CBAM Module을 제안한다. CBAM의 각 Gate와 Residual Block을 이용하여 각각의 표정에 대한 핵심 특징 정보들을 강조하여 Context 한 모델로 변화시켜주는 효과를 가지게 하였으며 Skip-Connection을 이용하여 기울기 소실 및 폭발에 강인하게 해주는 모듈을 제안한다. AI-HUB의 한국인 감정 인식을 위한 복합 영상 데이터 세트를 이용하여 총 6개의 클래스로 구분하였으며, F1-Score, Accuracy 기준으로 Identity-CBAM 모듈을 적용하였을 때 Vanilla ResNet50, ResNet101 대비 F1-Score 0.4~2.7%, Accuracy 0.18~2.03%의 성능 향상을 달성하였다. 또한, Guided Backpropagation과 Guided GradCam을 통해 시각화하였을 때 중요 특징점들을 더 세밀하게 표현하는 것을 확인하였다. 결과적으로 이미지 내 표정 분류 Task에서 Vanilla ResNet50, ResNet101을 사용하는 것보다 Identity-CBAM Module을 함께 사용하는 것이 더 적합함을 입증하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.679-682
• /
• 2012

본 논문은 지능형 실감 통신 서비스를 위해 단순히 영상 데이터만을 상대방에게 전송하는 것이나 영상 내에 포함된 사용자 상황 정보(위치, 얼굴, 문자 등)들을 인식한 후 자연스럽게 사용자 화면상에 오버레이 시키는 증강현실 기반의 상황인식 기술 개발을 위한 방안을 기술한다. 제안한 방안은 사용자 위치 인식 기술, 사용자 인식 기술, 웨어러블 컴퓨팅 기술, 증강현실 기술, 그리고 화질 개선을 위한 에러 은닉 기술들을 기반으로 한 연구이다. 제안한 방안은 향후 이동 통신 환경에서 사용자에게 보다 편리한 정보 제공 환경으로 공함으로써, 도서관이나 박물관 또는 관광지 등에서 사용자에게 쉽게 새로운 정보를 제공하는 수단으로 자리 잡을 수 있을 것이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.8
• /
• pp.1823-1830
• /
• 2003

This paper presents a multiple face detector based on a robust pupil detection technique. The pupil detector uses active illumination that exploits the retro­reflectivity property of eyes to facilitate detection. The detection range of this method is appropriate for interactive desktop and kiosk applications. Once the location of the pupil candidates are computed, the candidates are filtered and grouped into pairs that correspond to faces using heuristic rules. To demonstrate the robustness of the face detection technique, a dual mode face tracker was developed, which is initialized with the most salient detected face. Recursive estimators are used to guarantee the stability of the process and combine the measurements from the multi­face detector and a feature correlation tracker. The estimated position of the face is used to control a pan­tilt servo mechanism in real­time, that moves the camera to keep the tracked face always centered in the image.

• Journal of KIISE:Software and Applications
• /
• v.36 no.8
• /
• pp.673-682
• /
• 2009

Recently, photo sharing and publishing based Social Network Sites(SNSs) are increasingly attracting the attention of academic and industry researches. Unlike the face recognition environment addressed by existing works, face annotation problem under SNSs is differentiated in terms of daily updated images database, a limited number of training set and millions of users. Thus, conventional approach may not deal with these problems. In this paper, we proposed a face annotation method for sharing and publishing photographs that contain faces under a social network service using random projection, non-linear regression and representational state transfer. Our experiments on several databases show that the proposed method records an almost constant execution time with comparable accuracy of the PCA-SVM classifier.

• The KIPS Transactions:PartB
• /
• v.19B no.4
• /
• pp.225-230
• /
• 2012

In this paper, a blind person using a white cane as an adjunct of the things available sensing technology has been implemented. Sensing technology to implement things ultrasonic sensors and a webcam was used to process the data from the server computer. Ultrasonic sensors detect objects within 4meter people distinguish between those things that if the results based on the results will sound off. In this study, ultrasonic sensors, object recognition and human perception with the introduction of techniques and technologies developed for detecting objects in the lives of the visually impaired is expected to be greater usability.