• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• 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.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.811-812
• /
• 2018

딥러닝 기술의 발전에 따라, 딥러닝을 Human Pose Estimation에 적용하는 연구가 활발하게 수행되고 있다. 본 논문에서는 딥러닝을 이용한 여러 기술 중 가장 활발하게 사용이 되고 있는 Open Pose 와 Deeper Cut 기술의 특성을 분석한다.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.174-184
• /
• 2022

3D human pose estimation models can be classified into a multi-view model and a single-view model. In general, the multi-view model shows superior pose estimation performance compared to the single-view model. In the case of the single-view model, the improvement of the 3D pose estimation performance requires a large amount of training data. However, it is not easy to obtain annotations for training 3D pose estimation models. To address this problem, we propose a method to generate pseudo ground-truths of multi-view human pose data from a multi-view model and exploit the resultant pseudo ground-truths to train a single-view model. In addition, we propose a multi-view consistency loss function that considers the consistency of poses estimated from multi-view images, showing that the proposed loss helps the effective training of single-view models. Experiments using Human3.6M and MPI-INF-3DHP datasets show that the proposed method is effective for training single-view 3D human pose estimation models.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.4
• /
• pp.607-616
• /
• 2014

Face recognition is a big challenge in surveillance system since different rotation angles of the face make the difficulty to recognize the face of the same person. This paper proposes a novel method to recognize face with different head poses by using 3D information of the face. Firstly, head pose estimation (estimation of different head pose angles) is accomplished by the POSIT algorithm. Then, 3D face image data is constructed by using head pose estimation. After that, 2D image and the constructed 3D face matching is performed. Face verification is accomplished by using commercial face recognition SDK. Performance evaluation of the proposed method indicates that the error range of head pose estimation is below 10 degree and the matching rate is about 95%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.2
• /
• pp.469-478
• /
• 2010

This paper proposes the effective shape descriptor for 3D foot pose estimation. To reduce processing time, silhouette-based foot image database is built and meta information which involves the 3D pose of the foot is appended to the database. And we proposed a modified Centroid Contour Distance whose size of the feature space is small and performance of pose estimation is better than the others. In order to analyze performance of the descriptor, we evaluate time and spatial complexity with retrieval accuracy, and then compare with the previous methods. Experimental results show that the proposed descriptor is more effective than the previous methods on feature extraction time and pose estimation accuracy.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.2
• /
• pp.133-140
• /
• 2008

In this paper, a small and effective attitude estimation system for a humanoid robot was developed. Four small inertial sensors were packed and used for inertial measurements(3D accelerometer and three 1D gyroscopes.) An effective 3D pose estimation algorithm for low cost DSP using an extended Kalman filter was developed and evaluated. The 3D pose estimation algorithm has a very simple structure composed by 3 modules of a linear acceleration estimator, an external acceleration detector and an pseudo-accelerometer output estimator. The algorithm also has an effective switching structure based on probability and simple feedback loop for the extended Kalman filter. A special test equipment using linear motor for the testing of the 3D pose sensor was developed and the experimental results showed its very fast convergence to real values and effective responses. Popular DSP of TMS320F2812 was used to calculate robot's 3D attitude and translated acceleration, and the whole system were packed in a small size for humanoids robots. The output of the 3D sensors(pitch, roll, 3D linear acceleration, and 3D angular rate) can be transmitted to a humanoid robot at 200Hz frequency.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.277-278
• /
• 2021

최근 손동작 인식에 대한 연구들이 활발하다. 하지만 대부분 Depth 정보를 포함한3D 정보를 필요로 한다. 이는 기존 연구들이 Depth 카메라 없이는 동작하지 않는다는 한계점이 있다는 것을 의미한다. 본 프로젝트는 Depth 카메라를 사용하지 않고 2D 이미지에서 Hand Keypoint Detection을 통해 손동작 인식을 하는 방법론을 제안한다. 학습 데이터 셋으로 Facebook에서 제공하는 InterHand2.6M 데이터셋[1]을 사용한다. 제안 방법은 크게 두 단계로 진행된다. 첫째로, Object Detection으로 Hand Detection을 수행한다. 데이터 셋이 어두운 배경에서 촬영되어 실 사용 환경에서 Detection 성능이 나오지 않는 점을 해결하기 위한 이미지 합성 Augmentation 기법을 제안한다. 둘째로, Keypoint Detection으로 21개의 Hand Keypoint들을 얻는다. 실험을 통해 유의미한 벡터들을 생성한 뒤 Annoy (Approximate nearest neighbors Oh Yeah) Tree를 생성한다. 생성된 Annoy Tree들로 후처리 작업을 거친 뒤 최종 Pose Estimation을 완료한다. Annoy Tree를 사용한 Pose Estimation에서는 NN(Neural Network)을 사용한 것보다 빠르며 동등한 성능을 냈다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.1
• /
• pp.8-16
• /
• 2016

In the field of computer vision and robotics, bin picking is an important application area in which object pose estimation is necessary. Different approaches, such as 2D feature tracking and 3D surface reconstruction, have been introduced to estimate the object pose accurately. We propose a new approach where we can use both 2D image features and 3D surface information to identify the target object and estimate its pose accurately. First, we introduce a label detection technique using Maximally Stable Extremal Regions (MSERs) where the label detection results are used to identify the target objects separately. Then, the 2D image features on the detected label areas are utilized to generate 3D surface information. Finally, we calculate the 3D position and the orientation of the target objects using the information of the 3D surface.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.4
• /
• pp.23-35
• /
• 2012

This paper presents a 3-D pose (position and orientation) estimation method for an elliptic object in 3-D space. It is difficult to resolve the problem of determining 3-D pose parameters with respect to an elliptic feature in 3-D space by interpretation of its projected feature onto an image plane. As an alternative, we propose a two points-based pose estimation algorithm to seek the 3-D information of an elliptic feature. The proposed algorithm determines a homogeneous transformation uniquely for a given correspondence set of an ellipse and two coplanar points that are defined on model and image plane, respectively. For each plane, two triangular features are extracted from an ellipse and two points based on the polarity in 2-D projection space. A planar homography is first estimated by the triangular feature correspondences, then decomposed into 3-D pose parameters. The proposed method is evaluated through a series of experiments for analyzing the errors of 3-D pose estimation and the sensitivity with respect to point locations.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.328-335
• /
• 2022

With the advance of robot capabilities and functionalities, construction robots assisting construction workers have been increasingly deployed on construction sites to improve safety, efficiency and productivity. For close-proximity human-robot collaboration in construction sites, robots need to be aware of the context, especially construction worker's behavior, in real-time to avoid collision with workers. To recognize human behavior, most previous studies obtained 3D human poses using a single camera or an RGB-depth (RGB-D) camera. However, single-camera detection has limitations such as occlusions, detection failure, and sensor malfunction, and an RGB-D camera may suffer from interference from lighting conditions and surface material. To address these issues, this study proposes a novel method of 3D human pose estimation by extracting 2D location of each joint from multiple images captured at the same time from different viewpoints, fusing each joint's 2D locations, and estimating the 3D joint location. For higher accuracy, the probabilistic representation is used to extract the 2D location of the joints, considering each joint location extracted from images as a noisy partial observation. Then, this study estimates the 3D human pose by fusing the probabilistic 2D joint locations to maximize the likelihood. The proposed method was evaluated in both simulation and laboratory settings, and the results demonstrated the accuracy of estimation and the feasibility in practice. This study contributes to ensuring human safety in close-proximity human-robot collaboration by providing a novel method of 3D human pose estimation.