• The Journal of the Korea Contents Association
• /
• v.22 no.8
• /
• pp.73-80
• /
• 2022

Facial expression is an important means of representing characteristics in movies and animations, and facial capture technology can support the production of facial animation for 3D characters more quickly and effectively. Blendshape techniques are the most widely used methods for producing high-quality 3D face animations, but traditional blendshape often takes a long time to produce. Therefore, the purpose of this study is to achieve results that are not far behind the effectiveness of traditional production to reduce the production period of blend shape. In this paper, in order to make a blend shape, the method of using the cross-model to convey the blend shape is compared with the traditional method of making the blend shape, and the validity of the new method is verified. This study used kit boy developed by Unreal Engine as an experiment target conducted a facial capture test using two blend shape production techniques, and compared and analyzed the facial effects linked to blend shape.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.75-83
• /
• 2019

In this paper we present a new technique for simultaneously simplifying N triangule meshes with the same number of vertices and the same connectivities. Applying the existing simplification technique to each of the N triangule mesh creates a simplified mesh with the same number of vertices but different connectivities. These limits make it difficult to construct a simplified blend-shape model in a high-resolution blend-shape model. The technique presented in this paper takes into account the N meshes simultaneously and performs simplification by selecting an edge with minimal removal cost. Thus, the N simplified meshes generated as a result of the simplification retain the same number of vertices and the same connectivities. The efficiency and effectiveness of the proposed technique is demonstrated by applying simultaneous simplification technique to multiple triangle meshes.

• Proceedings of the Korea Contents Association Conference
• /
• 2014.11a
• /
• pp.169-170
• /
• 2014

컴퓨터 그래픽의 발달로 3D 애니메이션은 시각적 리얼리티와 화려한 영상미로 애니메이션 특유의 비현실적인 상황과 허구적 캐릭터가 주는 재미를 관객에게 전한다. 특히 캐릭터의 얼굴 표정은 관객과의 감정 소통과 의사전달에 중요한 정보로서 디테일한 연기를 필요로 한다. 이에 3D 애니메이션 캐릭터의 경우 페이셜에 다양한 기능들이 요구되며, 일반적인 블렌드 쉐입과 클러스터 외에도 만화적 표현을 위한 다양한 기술들이 사용된다. 기존의 공정 과정에는 한 페이셜에 이러한 모든 기능들이 접목되어 복잡하며 까다로운 페이셜 리깅 공정이 이뤄진다. 본 연구에서는 기존의 공정들에서 한정되게 사용되었던 블렌드 쉐입을 이용하여 다양한 기능들을 타겟팅하는 레이어 방식을 통해 효율적인 페이셜 리깅 공정을 연구하고자 한다.

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1279-1284
• /
• 2021

This paper addresses the task of transferring facial blendshape models to an arbitrary target face. Blendshape is a common method for the facial rig; however, production of blendshape rig is a time-consuming process in the current facial animation pipeline. We propose automatic blendshape facial rigging based on our blendshape transfer method. Our method computes the difference between source and target facial model and then transfers the source blendshape to the target face based on a deformation transfer algorithm. Our automatic method provides efficient production of a controllable digital human face; the results can be applied to various applications such as games, VR chating, and AI agent services.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.461-470
• /
• 2023

While speech animation generation employing deep learning has been actively researched for English, there has been no prior work for Korean. Given the fact, this paper for the very first time employs supervised deep learning to generate Korean speech animation. By doing so, we find out the significant effect of deep learning being able to make speech animation research come down to speech recognition research which is the predominating technique. Also, we study the way to make best use of the effect for Korean speech animation generation. The effect can contribute to efficiently and efficaciously revitalizing the recently inactive Korean speech animation research, by clarifying the top priority research target. This paper performs this process: (i) it chooses blendshape animation technique, (ii) implements the deep-learning model in the master-servant pipeline of the automatic speech recognition (ASR) module and the facial action coding (FAC) module, (iii) makes Korean speech facial motion capture dataset, (iv) prepares two comparison deep learning models (one model adopts the English ASR module, the other model adopts the Korean ASR module, however both models adopt the same basic structure for their FAC modules), and (v) train the FAC modules of both models dependently on their ASR modules. The user study demonstrates that the model which adopts the Korean ASR module and dependently trains its FAC module (getting 4.2/5.0 points) generates decisively much more natural Korean speech animations than the model which adopts the English ASR module and dependently trains its FAC module (getting 2.7/5.0 points). The result confirms the aforementioned effect showing that the quality of the Korean speech animation comes down to the accuracy of Korean ASR.

• Journal of the Korea Computer Graphics Society
• /
• v.12 no.2
• /
• pp.1-7
• /
• 2006

블렌드 쉐입 기반 얼굴 애니메이션을 위해 기저 모델(Expression basis)을 생성하는 방법을 크게 두 가지로 구분하면, 애니메이터가 직접 모델링을 하여 생성하는 방법과 통계적 방법에 기초하여 모델링하는 방법이 있다. 그 중 애니메이터에 의한 수동 모델링 방법으로 생성된 기저 모델은 직관적으로 표정을 인식할 수 있다는 장점으로 인해 전통적인 키프레임 제어가 가능하다. 하지만, 표정 공간(Expression Space)의 일부분만을 커버하기 때문에 모션데이터로부터의 재복원 과정에서 많은 오차를 가지게 된다. 반면, 통계적 방법을 기반으로 한 기저모델 생성 방법은 거의 모든 표정공간을 커버하는 고유 얼굴 모델(Eigen Faces)을 생성하므로 재복원 과정에서 최소의 오차를 가지지만, 시각적으로 직관적이지 않은 표정 모델을 만들어 낸다. 따라서 본 논문에서는 수동으로 생성한 기저모델을 유사 고유 얼굴 모델(Quasi-Eigen Faces)로 변형하는 방법을 제시하고자 한다. 결과로 생성되는 기저 모델은 시각적으로 직관적인 얼굴 표정을 유지하면서도 통계적 방법에 의한 얼굴표정 공간의 커버 영역과 유사하도록 확장할 수 있다.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.4
• /
• pp.1-11
• /
• 2022

The Skinned Multi-Person Linear Model (SMPL) is the most widely used parametric 3D Human Model optimized and learned from CAESAR, a 3D human scanned database created with measurements from 3,800 people living in United States in the 1990s. We point out the lack of racial diversity of body types in SMPL and propose K-SMPL that better represents Korean 3D body shapes. To this end, we develop a fitting algorithm to estimate 2,773 Korean 3D body shapes from Korean body measurement data. By conducting principle component analysis to the estimated Korean body shapes, we construct K-SMPL model that can generate various Korean body shape in 3D. K-SMPL model allows to improve the fitting accuracy over SMPL with respect to the Korean body measurement data. K-SMPL model can be widely used for avatar generation and human shape fitting for Korean.