• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.714-718
• /
• 2009

3D reconstruction of a human face from an image sequence remains an important problem in computer vision. We propose a method, based on a factorization algorithm, that reconstructs a 3D face model from short image sequences exhibiting rotational motion. Factorization algorithms can recover structure and motion simultaneously from one image sequence, but they usually require that all feature points be well tracked. Under rotational motion, however, feature tracking often fails due to occlusion and frame out of features. Additionally, the paucity of images may make feature tracking more difficult or decrease reconstruction accuracy. The proposed 3D reconstruction approach can handle short image sequences exhibiting rotational motion wherein feature points are likely to be missing. We implement the proposal as a reconstruction method; it employs image sequence division and a feature tracking method that uses Active Appearance Models to avoid the failure of feature tracking. Experiments conducted on an image sequence of a human face demonstrate the effectiveness of the proposed method.

• Archives of Reconstructive Microsurgery
• /
• 제9권2호
• /
• pp.172-178
• /
• 2000

With esthetic concern in the reconstruction of skin and soft tissue defects of face, the use of local flap has been the method of choice. However, when there is extensive tissue loss in the face, local flaps do not provide satisfactory results. The amazing development of microsurgical technique has decreased the percentage of free flap failure, thus making free flap use in reconstruction of facial soft tissue defects. Many free flaps has been applied for reconstruction of face defects. Especially, the radial forearm flap has numerous advantages with which facial reconstruction is made possible. But, its disadvantages are ; the sacrifice of one major artery supplying the hand and donor site complications. In order to circumvent these disadvantages, we employed posterior interosseous artery(PIA) forearm free flap for the reconstruction of the face defects. The posterior interosseous forearm island flap was first described by Zancolli and Angrigiani(1985). Currently, the PIA island flap and free flap have been used for hand reconstructions. The disadvantages of the PIA flap are ; the small caliber of the pedicle, different locations of the perforating branches, and the proximity of the motor branch of the radial nerve. But, its advantages lies in preserving the major artery of the hand, minimal donor site morbidity, and fairly well matched skin texture and color, and that the flap volume is sufficient, not too bulky with convenient handling. By using this flap, we performed 1 case of tumor resection and 1 case of traumatic defect. From our experiences we conclude that it is one of many useful methods in the reconstruction of the skin and soft tissue defects of the face. We also have discussed advantages and some limitations of various free flaps for reconstruction of the face.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권5호
• /
• pp.476-486
• /
• 2007

본 논문에서는 영역 기반 복원 방법을 통하여 한 장의 저해상도 얼굴 영상으로부터 고해상도 얼굴 영상을 복원하는 방법을 제안한다. 제안된 방법은 예제 기반 복원과 얼굴 영상을 형태 정보와 질감 정보로 나누어 표현하는 변형 가능 얼굴 모형에 기반한다. 먼저, 예제 기반 복원 방법의 성능을 개선하기 위하여, 전역 복원 결과와 국부적 복원 결과를 결합하는 영역 기반 복원 방법을 제안한다. 또한, 변형 가능 얼굴 모형의 장점을 해상도 복원에 적용하기 위하여, 확장된 변형 가능 얼굴 모형을 정의한다. 제안된 모형에서 얼굴 영상은 저해상도 얼굴 영상, 보간법을 통해 개선한 고해상도 얼굴 영상, 그리고 원래의 고해상도 얼굴 영상의 쌍으로 구성되며, 이는 다시 확장된 형태 정보와 확장된 질감 정보로 나뉜다. 다양한 실험을 통하여, 제안된 방법이 저해상도 얼굴 영상으로부터 고해상도 얼굴 영상을 효과적으로 복원함을 입증하였으며, 이 방법을 사용하여 원거리 감시 시스템에서 획득된 저해상도 얼굴 영상을 고해상도 얼굴 영상으로 합성함으로써, 얼굴 인식 시스템의 성능을 높일 수 있는 가능성을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.1157-1160
• /
• 2005

The denoising and reconstruction of color images are increasingly studied in the field of computer vision and image processing. Especially, the denoising and reconstruction of color face images are more difficult than those of natural images because of the structural characteristics of human faces as well as the subtleties of color interactions. In this paper, we propose a denoising method based on PCA reconstruction for removing complex color noises on human faces, which is not easy to remove by using vectorial color filters. The proposed method is composed of the following five steps; training of canonical eigenface space using PCA, automatic extracting of face features using active appearance model, relighing of reconstructed color image using bilateral filter, extraction of noise regions using the variance of training data, and reconstruction using partial information of input images (except the noise regions) and blending of the reconstructed image with the original image. Experimental results show that the proposed denosing method efficiently removes complex color noises on input face images.

• Journal of the Optical Society of Korea
• /
• 제18권5호
• /
• pp.500-506
• /
• 2014

A new 3-D reconstruction algorithm for the human face is proposed using the derivative Moir$\acute{e}$ topography which ensures fast and robust reconstruction even for rough surfaces. The Moir$\acute{e}$ interference fringe pattern is initially obtained through the projection Moir$\acute{e}$ topography based on phase shifting, and then differentiated to provide a full unwrapped phase map for a human face. $2{\pi}$ ambiguity, which has been a chronically unsolved problem with Moir$\acute{e}$ topography, is successfully surmounted by differentiating the Moir$\acute{e}$ fringe patterns both in x- and y-directions when the object is located in the x-y plane. A real human face is used for verifying the proposed derivative Moir$\acute{e}$ topography. A human face of 4 different phase-shifted images taken in the fixed plane is almost fully reconstructed in 3-D format in 0.1 mm lateral resolution.

• Applied Microscopy
• /
• 제46권2호
• /
• pp.100-104
• /
• 2016

Electron microscopy is currently the only available technique with a spatial resolution sufficient to identify fine neuronal processes and synaptic structures in densely packed neuropil. For large-scale volume reconstruction of neuronal connectivity, serial block-face scanning electron microscopy allows us to acquire thousands of serial images in an automated fashion and reconstruct neural circuits faster by reducing the alignment task. Here we introduce the whole reconstruction procedure of synaptic network in the rat hippocampal CA1 area and discuss technical issues to be resolved for improving image quality and segmentation. Compared to the serial section transmission electron microscopy, serial block-face scanning electron microscopy produced much reliable three-dimensional data sets and accelerated reconstruction by reducing the need of alignment and distortion adjustment. This approach will generate invaluable information on organizational features of our connectomes as well as diverse neurological disorders caused by synaptic impairments.

• 한국멀티미디어학회논문지
• /
• 제7권3호
• /
• pp.308-318
• /
• 2004

손상된 얼굴영상을 인식하는 문제는 자동 얼굴인식 시스템의 상용화에 중요한 논점이 되고 있다. 손상된 영상에서 얼굴을 인식하기 위한 방법은 크게 두 가지로 나누어진다. 첫째는 손상된 정보를 제거하여 인식을 하는 것이고, 둘째는 손상된 정보를 복원한 후에 인식을 하는 것이다. 본 논문에서는 손상된 영상을 코호넨 네트워크에 입력하여 손상된 영역을 추출한 다음에 주성분 분석을 통해 얻은 특징 벡터의 계수 추정을 이용하여 복원하는 방안을 제시한다. 본 논문에서 제안한 방법은 손상되지 않은 영역을 기반으로 하여 손상된 영역의 특징 벡터를 추정하고, 추정된 특징 벡터로써 손상된 영상을 복원하는 방법으로 학습되어지지 않은 영상에서도 복원이 가능하다. 본 연구실에서 얼굴 인식에 많은 영향을 미치는 눈과 입 부위를 인위적으로 손상시킨 영상을 실험하였을 때, 복원된 영상의 인식률은 학습된 영상의 경우에는 코호넨 네트워크를 사용한 방법과는 유사한 결과를 나타내고 대칭성을 이용한 방법에 비해서는 11.8%가 향상되었으며, 학습되지 않은 영상에서의 경우에는 코호넨 네트워크를 이용한 방법과 대칭성을 이용한 방법에 비해 각각 14%, 7% 향상되었다.

• Journal of International Society for Simulation Surgery
• /
• 제2권2호
• /
• pp.47-51
• /
• 2015

Faces play a key role in revealing the personalized attributes such as the identity, emotion, health condition, etc. Due to the importance of faces, computer-assisted face modeling and reconstruction have been actively studied both in computer vision and graphics community. Especially, face reconstruction and realistic face synthesis are well-grounded research problems and various approaches have been proposed during the last decade. In this paper, we discuss a wide range of existing work in face modeling by introducing their target applications, categorizing them upon their methodology and addressing their strength and weakness on performance. Finally, we introduce remaining research issues and suggest the future research direction in face modeling. We believe that this paper provides a high-level overview on face modeling techniques and helps understand the major research issues and the trends of methodology.

• Archives of Plastic Surgery
• /
• 제34권3호
• /
• pp.336-341
• /
• 2007

Purpose: The most accepted method for the reanimation of a paralyzed face is the two-stage method that combines cross-face nerve grafting with free-muscle transfer. Although the results of reconstruction with this method are satisfactory, there is an excessive delay between the stages, which prolongs the period of rehabilitation. In order to overcome this drawback, a one- stage, neurovascular free-flap reconstruction method using free neurovascular muscle flaps is introduced. Methods: From 1994 to 2004, 35 patients with longstanding facial palsy were treated. Fifteen patients underwent the single-stage reconstruction with the latissimus dorsi muscle, and 20 patients underwent the two-stage reconstruction method with the gracilis muscle. We compared the long-term results of the two methods of reconstruction. The mean follow-up period was 28.7 months for one-stage reconstruction, and 35.2 months for the two-stage, respectively. Results: In the patient group of the single stage reconstruction, both mouth corner excursion and animation grade were markedly improved at the final postoperative visit. Moreover, the first muscle contraction occurred earlier in this group, than in the two-stage reconstruction group. However, four patients in the single stage group never achieved a first muscle contraction or mouth corner excursion. Conclusion: Facial palsy is a very challenging condition for cosmetic surgeons to deal with. Traditional methods for treatment of chronic facial palsy use a two-stage muscle flap which is time-consuming and burdensome to patients, many of whom are averse to waiting 8 to 12 months to complete the two stages. The one-stage reconstruction method described herein uses a latissimus dorsi free-flap and has demonstrated consistent positive outcomes in clinical assessments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권7호
• /
• pp.2547-2567
• /
• 2021

It has been widely acknowledged that occlusion impairments adversely distress many face recognition algorithms' performance. Therefore, it is crucial to solving the problem of face image occlusion in face recognition. To solve the image occlusion problem in face recognition, this paper aims to automatically de-occlude the human face majority or discriminative regions to improve face recognition performance. To achieve this, we decompose the generative process into two key stages and employ a separate generative adversarial network (GAN)-based network in both stages. The first stage generates an initial coarse face image without an occlusion mask. The second stage refines the result from the first stage by forcing it closer to real face images or ground truth. To increase the performance and minimize the artifacts in the generated result, a new refine loss (e.g., reconstruction loss, perceptual loss, and adversarial loss) is used to determine all differences between the generated de-occluded face image and ground truth. Furthermore, we build occluded face images and corresponding occlusion-free face images dataset. We trained our model on this new dataset and later tested it on real-world face images. The experiment results (qualitative and quantitative) and the comparative study confirm the robustness and effectiveness of the proposed work in removing challenging occlusion masks with various structures, sizes, shapes, types, and positions.