• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.1048-1051
• /
• 2020

본 논문에서는 사람의 손에 관한 개별적으로 분리되어 진행되고 있는 손 위치 추정, 손 자세 추정, 손 동작 인식 작업을 통합하는 Faster-RCNN기반의 프레임워크를 제안하였다. 제안된 프레임워크에서는 RGB 동영상을 입력으로 하여, 먼저 손 위치에 대한 박스를 생성하고, 생성된 박스 정보를 기반으로 손 자세와 동작을 인식하도록 한다. 손 위치, 손 자세, 손 동작에 대한 정답을 동시에 모두 가지는 데이터셋이 존재하지 않기 때문에 Egohands, FPHA 데이터를 동시에 효과적으로 사용하는 방안을 제안하였으며 제안된 프레임워크를 FPHA데이터에 평가하였다., 손 위치 추정 정확도는 mAP 90.3을 기록했고, 손 동작 인식은 FPHA의 정답을 사용한 정확도에 근접한 70.6%를 기록하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.31-34
• /
• 2022

본 논문에서는 손목 부착형 웨어러블(Wearable) RGB 카메라를 통해 취득한 손 이미지에 최적화된 손 자세 추정모델과 학습방법을 제안한다. 최근 의료분야에서 활발하게 인공지능이 사용되고 있으며 그 중 이미지 인식을 중심으로 하는 진단 분야[1]가 괄목할만한 성과를 보인다. 본 연구에서는 웨어러블 카메라를 통해 얻은 손 자세를 활용하여 질병 진단에 적용할 계획이다. 또한, 본 연구수행을 통해 질병진단에 필요한 데이터 측정비용 절감 및 개인 맞춤형 진단서비스를 제공할 것으로 기대된다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.4
• /
• pp.469-475
• /
• 2015

This paper describes about the method for designing an extended Kalman filter to accurately measure the position of the spatial-phase system using a semiconductor sensor. Spatial position is expressed by the correlation of the rotated coordinate system attached to the body from the inertia coordinate system (a fixed coordinate system). To express the attitude, quaternion was adapted as a state variable, Then, the state changes were estimated from the input value which was measured in the gyro sensor. The observed data is the value obtained from the acceleration sensor. By matching between the measured value in the acceleration sensor and the predicted calculation value, the best variable was obtained. To increase the accuracy of estimation, designation of the extended Kalman filter was performed, which showed excellent ability to adjust the estimation period relative to the sensor property. As a result, when a three-axis gyro sensor and a three-axis acceleration sensor were adapted in the estimator, the RMS(Root Mean Square) estimation error in simulation was retained less than 1.7[$^{\circ}$], and the estimator displayed good property on the prediction of the state in 100 ms measurement period.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.2
• /
• pp.304-310
• /
• 2010

Chronic pain and herniated disk is a common disease that 80% of adults are experienced. There diseases rates of caused by the physical shock, such as the traffic accident, and the accidental fall is about 10%. And the most of these diseases is caused by having habitual incorrect position. People know that incorrect position would cause to accumulate continuous stress, but it is not easy to correct position. Because it does not recognize incorrect position repeated habitual consequently. This system collects data of user position after sensors that could measure position attach on use and presumes correct position used by position presumption algorithms. Its system purpose is continuing incorrect position could be aware to user and lead to change to correct position to prevent habituation of incorrect position. If habitual of correct position continues through accurate measurement and repeating cognitive learning, it would help for children and chronic patience.

• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.41-51
• /
• 2018

Recently, in the field of video surveillance, deep learning based learning method is applied to intelligent video surveillance system, and various events such as crime, fire, and abnormal phenomenon can be robustly detected. However, since occlusion occurs due to the loss of 3d information generated by projecting the 3d real-world in 2d image, it is need to consider the occlusion problem in order to accurately detect the object and to estimate the pose. Therefore, in this paper, we detect moving objects by solving the occlusion problem of object detection process by adding depth information to existing RGB information. Then, using the convolution neural network in the detected region, the positions of the 14 keypoints of the human joint region can be predicted. Finally, in order to solve the self-occlusion problem occurring in the pose estimation process, the method for 3d human pose estimation is described by extending the range of estimation to the 3d space using the predicted result of 2d keypoint and the deep neural network. In the future, the result of 2d and 3d pose estimation of this research can be used as easy data for future human behavior recognition and contribute to the development of industrial technology.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.76-77
• /
• 2022

In this paper, we propose a method to stabilize the 3D pose estimation result of a 3D volumetric data sequence by matching the pose estimation results from multi-view. Draw a circle centered on the volumetric model and project the model from the viewpoint at regular intervals. After performing Openpose 2D pose estimation on the projected 2D image, the 2D joint is matched to localize the 3D joint position. The tremor of 3D joints sequence according to the angular spacing was quantified and expressed in graphs, and the minimum conditions for stable results are suggested.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.8
• /
• pp.63-70
• /
• 2006

Due to the small size and weight of micro aerial vehicle (MAV), only miniaturized MEMS type sensors are applicable for MAV autonomous flight system. In this paper, we propose a accelerometer and gyro mixing algorithm to improve an attitude performance of MEMS type sensors. The performance of the proposed mixing algorithm is compared with the performance of fuzzy-based mixing algorithm through simulation. The simulation results show that the attitude compensation method through the attitude compensation has better performance than the fuzzy-based mixing method for MAV attitude estimation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.811-813
• /
• 2005

주어진 영상에서 인체를 찾고 그 자세를 인식하기 위해 자세나 조영 조건의 변화에 됨 민감한 방법으로 구성요소에 기반한 접근이 있다. 본 논문에서는 10개의 구성요소와 그들간의 유연한 연결로 구성된 인체모델을 사용한다. 각 구성요소는 기하학적, 명시적, 다른 구성요소와의 연결요소에 대한 정보로 구성되어 있다. 인체구성요소 사이의 계층적 연결은 일반-상세 탐색으로 시간효율적인 인체 매칭을 가능케 한다. 본 논문에서는 새로운 인체를 찾을 때마다 인체 구성요소를 갱신함으로써 자세 및 조명 변화에 보다 적응적으로 자세를 추정하는 방법을 제안한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.509-516
• /
• 2022

Accurate pose prediction of objects in 3D space is an important visual recognition technique widely used in many applications such as scene understanding in both indoor and outdoor environments, robotic object manipulation, autonomous driving, and augmented reality. Most previous works for object pose estimation have the limitation that they require an exact 3D CAD model for each object. Unlike such previous works, this paper proposes a novel neural network model that can predict the poses of unknown objects based on only their RGB color images without the corresponding 3D CAD models. The proposed model can obtain depth maps required for unknown object pose prediction by using an adaptive depth estimator, AdaBins,. In this paper, we evaluate the usefulness and the performance of the proposed model through experiments using benchmark datasets.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.106-114
• /
• 2017

In this paper, we propose the method of designing an extended kalman filter in order to accurately measure the position of the spatial-phase system using sensor fusion. We use the quaternion as a state variable in expressing the attitude of an object. Then, the attitude of rigid body can be calculated from the accelerometer and magnetometer by applying the Gauss-Newton method. We estimate the changes of state by using the measurements obtained from the gyroscope, the quaternion, and the vision informations by ARVR_SDK. To increase the accuracy of estimation, we designed and implemented the extended kalman filter, which showed excellent ability to adjust and compensate the sensor error. As a result, we could experimentally demonstrate that the reliability of the attitude estimation value can be significantly increased.