• 로봇학회논문지
• /
• 제11권4호
• /
• pp.226-234
• /
• 2016

In this study, a model-referenced underwater navigation algorithm is proposed for high-precise underwater navigation using monocular vision near underwater structures. The main idea of this navigation algorithm is that a 3D model-based pose estimation is combined with the inertial navigation using an extended Kalman filter (EKF). The spatial information obtained from the navigation algorithm is utilized for enabling the underwater robot to navigate near underwater structures whose geometric models are known a priori. For investigating the performance of the proposed approach the model-referenced navigation algorithm was applied to an underwater robot and a set of experiments was carried out in a water tank.

• 대한전자공학회논문지SP
• /
• 제40권6호
• /
• pp.31-40
• /
• 2003

대부분의 얼굴인식 시스템은 현재 2차원 영상을 기반으로 많은 분야에 응용되고 있다. 그러나 2차원 얼굴인식 시스템은 심하게 변화된 얼굴 포즈에 강인한 얼굴인식이 매우 어렵다. 이에 얼굴 포즈 추정은 정면 영상이 아닐 경우 인식률 향상을 위한 필수적인 과정이라 할 수 있다. 그러므로, 본 논문은 3차원 얼굴인식을 위한 새로운 얼굴 포즈 추정 방식을 제안한다 먼저 3차원 거리(range) 영상이 입력될 때 얼굴 곡선에 기반한 자동 얼굴 특징점 추출 기법을 적용한다. 추출된 특징점을 바탕으로 오류 보상 특이치 분해를 적용 한 새로운 3차원 얼굴 포즈 추정 방식을 제안한다. 특이치 분해를 이용하여 초기 회전각을 획득한 후 존재하는 오류를 보다 세밀하게 보상한다. 제안 알고리즘은 정규화된 3차원 얼굴 공간에서 추출된 특징점의 기하학적 위치를 이용하여 수행된다. 또한 3차원 얼굴인식을 위하여 3차원 최근접 이웃 분류기를 이용한 데이터베이스내에서 후보 얼굴을 선택하는 방식을 제안한다. 실험 결과를 통해 다양한 얼굴 포즈에 대하여 제안 알고리즘의 효율성과 타당성을 검증하였다.

• 한국멀티미디어학회논문지
• /
• 제21권10호
• /
• pp.1162-1170
• /
• 2018

This paper presents an augmented reality system for multiple users. The proposed system performs ar image-based pose estimation of users and pose of each user is shared with other uses via a network server. For camera-based pose estimation, we install multiple markers in a pre-determined space and select the marker with the best appearance. The marker is detected by corner point detection and for robust pose estimation. the marker's corner points are tracked by optical flow tracking algorithm. Experimental results show that the proposed system successfully provides an augmented reality application to multiple users even when users are rapidly moving and some of markers are occluded by users.

• 대한임베디드공학회논문지
• /
• 제18권2호
• /
• pp.41-50
• /
• 2023

Gait analysis is an important tool in the clinical management of cerebral palsy, allowing for the assessment of condition severity, identification of potential gait abnormalities, planning and evaluation of interventions, and providing a baseline for future comparisons. However, traditional methods of gait analysis are costly and time-consuming, leading to a need for a more convenient and continuous method. This paper proposes a method for analyzing the posture of cerebral palsy patients using only smartphone videos and deep learning models, including a ResNet-based image tilt correction, AlphaPose for human pose estimation, and SmoothNet for temporal smoothing. The indicators employed in medical practice, such as the imbalance angles of shoulder and pelvis and the joint angles of spine-thighs, knees and ankles, were precisely examined. The proposed system surpassed pose estimation alone, reducing the mean absolute error for imbalance angles in frontal videos from 4.196° to 2.971° and for joint angles in sagittal videos from 5.889° to 5.442°.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.5-7
• /
• 2021

자세 추정을 위한 모션 캡처 데이터 파일에는 주변 환경과 움직임의 정도에 따라 부정확한 데이터가 존재할 수 있으므로, 이를 보정하는 작업이 필요하다. 기존에는 직접 후처리 과정을 통해 부정확한 데이터를 복원하였으나, 최근에는 자동화된 방법으로 LSTM, R-CNN 등 다양한 종류의 신경망을 사용한다. 하지만 신경망 기반의 데이터 복원 방법들은 컴퓨터 자원을 많이 요구하므로, 본 논문에서는 신경망 기반의 방법보다 자원 사용량은 낮추면서 데이터 복원율은 유지하는 방법을 제안한다. 제안하는 방법은 자세 측정 데이터(c3d)를 활용하여 부정확한 자세 데이터를 자동으로 복원한다. 실험 결과, 데이터의 부정확한 정도에 따라 89%에서부터 99% 정도의 데이터 복원율을 보였다.

• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.659-666
• /
• 2021

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• pp.2388-2398
• /
• 2017

In this study, we propose a robust face recognition system to pose variations based on automatic pose estimation. Radial basis function neural network is applied as one of the functional components of the overall face recognition system. The proposed system consists of preprocessing and recognition modules to provide a solution to pose variation and high-dimensional pattern recognition problems. In the preprocessing part, principal component analysis (PCA) and 2-dimensional 2-directional PCA ($(2D)^2$ PCA) are applied. These functional modules are useful in reducing dimensionality of the feature space. The proposed RBFNNs architecture consists of three functional modules such as condition, conclusion and inference phase realized in terms of fuzzy "if-then" rules. In the condition phase of fuzzy rules, the input space is partitioned with the use of fuzzy clustering realized by the Fuzzy C-Means (FCM) algorithm. In conclusion phase of rules, the connections (weights) are realized through four types of polynomials such as constant, linear, quadratic and modified quadratic. The coefficients of the RBFNNs model are obtained by fuzzy inference method constituting the inference phase of fuzzy rules. The essential design parameters (such as the number of nodes, and fuzzification coefficient) of the networks are optimized with the aid of Particle Swarm Optimization (PSO). Experimental results completed on standard face database -Honda/UCSD, Cambridge Head pose, and IC&CI databases demonstrate the effectiveness and efficiency of face recognition system compared with other studies.

• 인터넷정보학회논문지
• /
• 제19권6호
• /
• pp.41-51
• /
• 2018

최근 영상 감시 분야에서는 지능형 영상 감시 시스템에 딥 러닝 기반 학습 방법이 적용되어 범죄, 화재, 이상 현상과 같은 다양한 이벤트들을 강건하게 탐지 할 수 있게 되었다. 그러나 3차원 실세계를 2차원 영상으로 투영시키면서 발생하는 3차원 정보의 손실로 인하여 폐색 문제가 발생하기 때문에 올바르게 객체를 탐지하고, 자세를 추정하기 위해서는 폐색 문제를 고려하는 것이 필요하다. 따라서 본 연구에서는 기존 RGB 정보에 깊이 정보를 추가하여 객체 탐지 과정에서 나타나는 폐색 문제를 해결하여 움직이는 객체를 탐지하고, 탐지된 영역에서 컨볼루션 신경망을 이용하여 인간의 관절 부위인 14개의 키포인트의 위치를 예측한다. 그 다음 자세 추정 과정에서 발생하는 자가 폐색 문제를 해결하기 위하여 2차원 키포인트 예측 결과와 심층 신경망을 이용하여 자세 추정의 범위를 3차원 공간상으로 확장함으로써 3차원 인간 자세 추정 방법을 설명한다. 향후, 본 연구의 2차원 및 3차원 자세 추정 결과는 인간 행위 인식을 위한 용이한 데이터로 사용되어 산업 기술 발달에 기여 할 수 있다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.735-739
• /
• 2005

This paper presents a new system to estimate the head pose of human in interactive indoor environment that has dynamic illumination change and large working space. The main idea of this system is to suggest a new morphological feature for estimating head angle from stereo disparity map. When a disparity map is obtained from stereo camera, the matching confidence value can be derived by measurements of correlation of the stereo images. Applying a threshold to the confidence value, we also obtain the specific morphology of the disparity map. Therefore, we can obtain the morphological shape of disparity map. Through the analysis of this morphological property, the head pose can be estimated. It is simple and fast algorithm in comparison with other algorithm which apply facial template, 2D, 3D models and optical flow method. Our system can automatically segment and estimate head pose in a wide range of head motion without manual initialization like other optical flow system. As the result of experiments, we obtained the reliable head orientation data under the real-time performance.

• 제어로봇시스템학회논문지
• /
• 제12권5호
• /
• pp.450-456
• /
• 2006

Monte Carlo localization is known to be one of the most reliable methods for pose estimation of a mobile robot. Although MCL is capable of estimating the robot pose even for a completely unknown initial pose in the known environment, it takes considerable time to give an initial pose estimate because the number of random samples is usually very large especially for a large-scale environment. For practical implementation of MCL, therefore, a reduction in sample size is desirable. This paper presents a novel approach to reducing the number of samples used in the particle filter for efficient implementation of MCL. To this end, the topological information generated through the thinning technique, which is commonly used in image processing, is employed. The global topological map is first created from the given grid map for the environment. The robot then scans the local environment using a laser rangefinder and generates a local topological map. The robot then navigates only on this local topological edge, which is likely to be similar to the one obtained off-line from the given grid map. Random samples are drawn near the topological edge instead of being taken with uniform distribution all over the environment, since the robot traverses along the edge. Experimental results using the proposed method show that the number of samples can be reduced considerably, and the time required for robot pose estimation can also be substantially decreased without adverse effects on the performance of MCL.