• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2009년도 학술대회
• /
• pp.186-192
• /
• 2009

본 논문은 비전을 이용한 인간 정면 상반신 포즈를 인식 방법에 대해서 기술한다. 일반적으로 HCI(Human Computer Interaction)와 HRI(Human Robot Interaction)에서는 인간이 정면을 바라볼 때 얼굴, 손짓으로 의사소통 하는 경우가 많기 때문에 본 논문에서는 인식의 범위를 인간의 정면 그리고 상반신에 대해서만 한정한다. 인간 포즈인식의 주요 두 가지 어려움은 첫째 인간은 다양한 관절로 이루어진 객체이기 때문에 포즈의 자유도가 높은 문제점 때문에 모델링이 어렵다는 것이다. 둘째는 모델링된 정보와 영상과의 매칭이 어려운 것이다. 이를 해결하기 위해 본 논문에서는 모델링이 쉬운 그림모델(Pictorial Model)을 이용해 인체를 다수 사각형 파트로 모델링 하였고 이를 이용해 주요한 상반신 포즈를 DB화 해 인식한다. DB 포즈로 표현되지 못하는 세부포즈는 인식된 주요 포즈 파라미터로 부터 파티클필터를 이용해 예측한 다수 파티클로부터 가장 높은 사후분포를 갖는 파티클을 찾아 주요 포즈를 업데이트하여 결정한다. 따라서 주요한 포즈 인식과 이를 기반으로 한 세부 포즈를 추적하는 두 단계를 통해 인체 정면 상반신 포즈를 정확하게 인식 할 수 있다.

• Advances in robotics research
• /
• 제2권2호
• /
• pp.129-140
• /
• 2018

With the increasing demand for the development of human pose estimation, such as human-computer interaction and human activity recognition, there have been numerous approaches to detect the 2D poses of people in images more efficiently. Despite many years of human pose estimation research, the estimation of human poses with images remains difficult to produce satisfactory results. In this study, we propose a robust 2D human body pose estimation method using an RGB camera sensor. Our pose estimation method is efficient and cost-effective since the use of RGB camera sensor is economically beneficial compared to more commonly used high-priced sensors. For the estimation of upper-body joint positions, semantic segmentation with a fully convolutional network was exploited. From acquired RGB images, joint heatmaps accurately estimate the coordinates of the location of each joint. The network architecture was designed to learn and detect the locations of joints via the sequential prediction processing method. Our proposed method was tested and validated for efficient estimation of the human upper-body pose. The obtained results reveal the potential of a simple RGB camera sensor for human pose estimation applications.

• 정보처리학회논문지B
• /
• 제18B권1호
• /
• pp.11-20
• /
• 2011

Pictorial Structures(PS)는 동적 프로그래밍을 이용하여 인체의 포즈 추적 및 인식 하는 것에 매우 효과적인 방법으로 알려져 있다. 본 논문에서 상반신 포즈는 PS와 Particle filter(PF)에 의한 동적 프로그래밍 기법으로 추적된다. PF와 같은 동적프로그래밍에서 마코프 연쇄 (Markov Chain) 기반 동적 움직임 모델은 높은 자유도를 갖는 상반신 포즈를 예측하기 어려운 단점이 있다. 본 논문에서 제안하는 방법은 키포즈 기반 예측분포이며, 이것은 상반신 실루엣과 키포즈(Key Pose)들 사이의 유사도를 참고하여 파티클(Particle)을 적절히 예측하는 것이다. 실험 결과를 통해 제안된 방법은 기존 방법 성능을 70.51% 향상시킨 것을 확인하였다.

• 한국콘텐츠학회논문지
• /
• 제11권12호
• /
• pp.619-628
• /
• 2011

일반적으로 비전기반 3차원 인체 포즈 인식 기술은 HCI(Human-Computer Interaction)에서 인간의 제스처를 전달하기 위한 방법으로 사용된다. 특수한 환경에서 단순한 2차원 움직임 포즈만 인식할 수 있는 2차원 포즈모델 기반 인식 방법에 비해 3차원 관절을 묘사한 포즈모델은 관절각에 대한 정보와 신체 부위의 모양정보를 선행지식으로 사용할 수 있어서 좀 더 일반적인 환경에서 복잡한 3차원 포즈도 인식할 수 있다는 장점이 있다. 이 논문은 인체의 3차원 관절 정보를 이용한 포즈 인식 기술을 인터페이스로 활용한 상호작용 게임 콘텐츠 개발에 관해 기술한다. 제안된 시스템에서 사용되는 포즈는 인체 관절 중 14개 관절의 3차원 위치정보를 이용해서 구성한 포즈 템플릿과 현재 사용자의 포즈를 비교해 인식된다. 이 방법을 이용하여 제작된 시스템은 사용자가 부가적인 장치의 사용 없이 사용자의 몸동작만으로 자연스럽게 게임 콘텐츠를 조작할 수 있도록 해준다. 제안된 3차원 인식 기술을 게임 콘텐츠에 적용하여 성능을 평가한다. 향후 다양한 환경에서 더욱 강건하게 포즈를 인식할 수 있는 연구를 수행할 계획이다.

• 한국가정과학회지
• /
• 제7권3호
• /
• pp.15-33
• /
• 2004

A precise understanding of the human form in static pose serves as the basis of designing clothing. When the human body is in motion, however, even an article of clothing designed to fit the human form in static pose can pull and change, thus restricting the body. In order to increase the fit of the clothing, which may be termed the second skin, its form and measurements therefore must be determined in correlation not only with the formal characteristics of the human body, in static pose but also with its functional characteristics in motion, as caused by the movements of the human body. In this study, the motion factor was selected as the primary basis for designing slacks with good fit in both static and moving states. By indentifying the areas in which lower limb movement cause significant changes in body surface lines, we suggest several application methods for designing slacks. Using unmarried female university students aged 18 - 24 as subjects, a total of 32 body surface categories (15 body surface lines and 17 body surface segment lines) were measured in one static and 9 movement poses. In particular, expansion and contraction levels and rates were measured and used in the analysis. The analysis first involved the calculation of the average measurement per body part in body surface line in static pose as well as of the average expansion and contraction levels and rates in 9 lower limb movements. Two-way MANOVA and multiple comparison analysis (Tukey) were conducted on movements and individual somatotypes regarding measurement per body part and expansion and contraction rates. Body parts whose measurements of body surface lines differed significantly in body surface line in static pose versus in movement were then identified. The results of this study are as follows. First, changes in body surface lines caused by lower limb movements were significant in all body surface lines of the lower trunk, both horizontal and vertical, with the exception of abdomen girth, midway thigh girth, ankle girth, hip length, and posterior knee girth. Second, significantly expanded 10 body surface lines in moving pose were detected and illustrated in table 4. These body parts should be studied in designing or pattern designing, especially for close-fitting pants, in using stretch fabric, and in sensory evaluation of good fit during movement.

• 한국의류산업학회지
• /
• 제4권6호
• /
• pp.544-549
• /
• 2002

By the movable thermal manikin developed by China Dong Hua university, the laws of clothing thermal insulation influenced by dressing poses are studied. It is found that $I_a$ on nude thermal manikin has no relation to testing pose as a whole (notable level is 5%), while the change of testing pose influences $I_a$ value on parts of body obviously. The testing result $I_{cle}$ on clothed thermal manikin has relation to testing pose. The $I_{cle}$ value of the whole body in seated pose decreases 20 percent compared with that in standing pose (notable level is 1%). In view of heat transmission theory, the reasons are pointed out based on the knowledge of heat transmission.

• 한국의류학회지
• /
• 제38권5호
• /
• pp.769-782
• /
• 2014

This study considers the correlation between model pose and clothes in Patrick Demarchelier's fashion photos as well as expression characteristics. The conclusions of the study are as follows. The type of model pose in Patrick Demarchelier's fashion photos can be categorized into five types: maximized type of upper and lower body part, minimized type of upper and lower body part, maximized type of upper body and minimized type of lower body part, minimized type of upper body and maximized type of lower body part, basic type of upper and lower body part. In case of having examined the correlation between body movement and costume, the clothes in the model pose included in the maximization of the body were formed mainstream by silhouette, which was formed by decorative elements or full drapery. In the model poses included in the reduction of the body, the costume tended to expose many parts of the body to provide a simple or structural silhouette form. The costume was expressed in colorful form and poses assumed without body movements. The expression characteristics of the model poses in Patrick Demarchelier's fashion photos were sensuality, dynamicity, and simplicity. First, sensuality was expressed as feminine sensuality accompanying an erotic mood by naturally emphasizing a woman's breast or leg by reducing the body. Second, dynamicity provided a vividness to the image as if directly living and moving by highlighting the rhythmic aspect of the body. Simplicity aroused the effect of paying attention to clothes or other incidental elements rather than the image expressed by the body of a model by excluding body movement.

• International journal of advanced smart convergence
• /
• 제7권2호
• /
• pp.95-100
• /
• 2018

In this paper, we propose a binarized multi-scale module to accelerate the speed of the pose estimating deep neural network. Recently, deep learning is also used for fine-tuned tasks such as pose estimation. One of the best performing pose estimation methods is based on the usage of two neural networks where one computes the heat maps of the body parts and the other computes the part affinity fields between the body parts. However, the convolution filtering with a large kernel filter takes much time in this model. To accelerate the speed in this model, we propose to change the large kernel filters with binarized multi-scale modules. The large receptive field is captured by the multi-scale structure which also prevents the dropdown of the accuracy in the binarized module. The computation cost and number of parameters becomes small which results in increased speed performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.800-816
• /
• 2018

In recent years, video surveillance research has been able to recognize various behaviors of pedestrians and analyze the overall situation of objects by combining image analysis technology and deep learning method. Human Activity Recognition (HAR), which is important issue in video surveillance research, is a field to detect abnormal behavior of pedestrians in CCTV environment. In order to recognize human behavior, it is necessary to detect the human in the image and to estimate the pose from the detected human. In this paper, we propose a novel approach for 2D Human Pose Estimation based on object detection using RGB-D information. By adding depth information to the RGB information that has some limitation in detecting object due to lack of topological information, we can improve the detecting accuracy. Subsequently, the rescaled region of the detected object is applied to ConVol.utional Pose Machines (CPM) which is a sequential prediction structure based on ConVol.utional Neural Network. We utilize CPM to generate belief maps to predict the positions of keypoint representing human body parts and to estimate human pose by detecting 14 key body points. From the experimental results, we can prove that the proposed method detects target objects robustly in occlusion. It is also possible to perform 2D human pose estimation by providing an accurately detected region as an input of the CPM. As for the future work, we will estimate the 3D human pose by mapping the 2D coordinate information on the body part onto the 3D space. Consequently, we can provide useful human behavior information in the research of HAR.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권3호
• /
• pp.209-218
• /
• 2013

인간의 동작 인식에 대한 이전 연구는 주로 관절체로 표현된 신체 움직임을 추적하고 분류하는데 초점을 맞춰 왔다. 이 방식들은 실제 이미지 사용 환경에서 신체 부위에 대한 정확한 분류가 필요하다는 점이 까다롭기 때문에 최근의 동작 인식 연구 동향은 시공간상의 관심 점과 같이 저수준의, 더 추상적인 외형특징을 이용하는 방식이 일반화되었다. 하지만 몇 년 사이 자세 예측 기술이 발전하면서 자세 기반 방식에 대한 시각을 재정립하는 것이 필요하다. 본 연구는 외형 기반 방식에서 저수준의 외형특징만으로 분류기를 학습시키는 것이 충분한지에 대한 문제를 제기하면서 자세 예측을 이용한 효과적인 자세기반 동작인식 방식을 제안하였다. 이를 위해 다양한 감정을 표현하는 동작 시나리오를 대상으로 외형 기반, 자세 기반 특징 및 두 가지 특징을 조합한 방식을 비교하였다. 실험 결과, 자세 예측을 이용한 자세 기반 방식이 저수준의 외형특징을 이용한 방식보다 감정 동작 분류 및 인식 성능이 더 나았으며 잡음 때문에 심하게 망가진 이미지의 감정 동작 인식에도 자세 예측을 이용한 자세기반의 방식이 효과적이었다.