• 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 Broadcast Engineering
• /
• v.27 no.6
• /
• pp.944-947
• /
• 2022

Due to the coronavirus pandemic, the wearing of a mask has been increasing worldwide; thus, the importance of image analysis on masked face images has become essential. Although head pose estimation can be applied to various face-related applications including driver attention, face frontalization, and gaze detection, few studies have been conducted to address the performance degradation caused by masked faces. This study proposes a new data augmentation that synthesizes the masked face, depending on the face image size and poses, which shows robust performance on BIWI benchmark dataset regardless of mask-wearing. Since the proposed scheme is not limited to the specific model, it can be utilized in various head pose estimation models.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.11
• /
• pp.48-56
• /
• 2005

Predictive filter theory proposed recently can be characterized by inherent advantages of estimating modelling error and overcoming the disadvantage of the Kalman filter theory. A one-step ahead error is minimized to produce optimized filter performance in the form of the predictive filter. The main advantage of this filter lies in the ability to estimate both state vector and system model error. In this paper, attitude estimation results based upon the predictive filter theory is addressed. Mathematical formulation for estimating bias signal is peformed by using the predictive filter theory, and attitude estimation based upon vector observation is presented. From the results of this study, the potential applicability of the predictive filter is highlighted.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.68-71
• /
• 2020

최근 딥러닝을 이용한 인체 자세 추정(human pose estimation) 연구가 활발히 진행되고 있다. 그 중 구조가 간단하면서도 성능이 강력하여 널리 사용되고 있는 딥러닝 네트워크 모델은 이미지 분류(image classification)에 사용되는 백본 네트워크(backbone network)와 디컨볼루션 출력망(deconvolution head network)을 이어 붙인 구조를 갖는다[1]. 기존의 디컨볼루션 출력망은 디컨볼루션 층을 쌓아 낮은 해상도의 특징맵을 모두 높은 해상도로 변환한 후 최종 인체 자세 추정을 하는데 이는 다양한 해상도에서 얻어낸 특징들을 골고루 활용하기 힘들다는 단점이 있다. 따라서 본 논문에서는 매 디컨볼루션 층 이후에 인체 자세 추정을 하여 다양한 해상도에서 연산을 하고 이를 종합하여 최종 인체 자세 추정을 하는 방법을 제안한다. 실험 결과 Res50 과 기존의 디컨볼루션 출력망의 경우 0.717 AP 를 얻었는데 Res101 과 기존의 디컨볼루션 출력망을 사용한 결과 50% 이상의 파라미터 수 증가와 함께 0.727 AP, 즉 0.010AP 의 성능 향상이 이루어졌다. 이에 반해 Res50 에 다중 해상도 디컨볼루션 출력망을 사용한 결과 약 1%의 파라미터 수 증가 만으로 0.720 AP, 즉 0.003 AP 의 성능 향상이 이루어졌다. 이를 통해 디컨볼루션 출력망 구조를 개선하면 매우 적은 파라미터 수 증가 만으로도 인체 자세 추정의 성능을 효과적으로 향상시킬 수 있음을 확인하였다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.8
• /
• pp.783-789
• /
• 2017

In terms of 3D orientation estimation based on nine-axis IMMU(inertial and magnetic measurement unit), there are two disturbance components decreasing estimation accuracy: one is external acceleration disturbing accelerometer's signals and the other is magnetic disturbance related to magnetometer's signals. In order to minimize effects by these two disturbances, two approaches including switching approach and model-based approach have been suggested and further research comparing these two has also been conducted. Nevertheless, effect of disturbance modeling differences on orientation estimation accuracy in model-based approach has not been studied before. This paper compares the recently reported two orientation estimation algorithms that have difference in disturbance models, in order to investigate the effect of disturbance models on accuracy of IMMU-based orientation estimation under various operating conditions. This research shows that the difference in disturbance models leads to difference in process noise covariance matrix. Consequently, this affected the orientation estimation, i.e., the estimation differences between the algorithms were root mean square errors of $1.35^{\circ}$ in average and $3.63^{\circ}$ in yaw estimation.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.69-75
• /
• 2023

In recent years, there has been a growing interest in deep learning-based animal pose estimation, especially in the areas of animal behavior analysis and healthcare. However, existing animal pose estimation techniques do not perform well when body parts are occluded or not present. In particular, the occlusion of dog tail or ear might lead to a significant degradation of performance in pet behavior and emotion recognition. In this paper, to solve this intractable problem, we propose a simple yet novel framework for pet pose estimation where pet pose is predicted on an outpainted image where some body parts hidden outside the input image are reconstructed by the image inpainting network preceding the pose estimation network, and we performed a preliminary study to test the feasibility of the proposed approach. We assessed CE-GAN and BAT-Fill for image outpainting, and evaluated SimpleBaseline for pet pose estimation. Our experimental results show that pet pose estimation on outpainted images generated using BAT-Fill outperforms the existing methods of pose estimation on outpainting-less input image.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.7-14
• /
• 2006

본 논문에서는 3D 복원과 카메라 측정과정 없이 정확하게 카메라 자세를 계산하고 가상객체를 비디오에 합성하기 위한 단일 프레임 기반의 고속 계산 기법을 제안한다. 객체의 로컬 좌표와 단일 이미지에서의 대응되는 이미지 좌표로부터 카메라 자세를 계산한다. 정사영 투영모델에서의 분해기법에 기반한 구조 계산 방법으로 카메라 자세의 고속 추정이 가능하다. 정사영 투영모델에 기반하기 때문에 참조점의 설정에 따라 정확도가 달라진다. 객체에 따라 참조점을 설정하여 정확한 카메라 자세를 계산하는 방법을 제안한다. 카메라 자세 및 물체의 형태는 단일 프레임 기반으로 수행되며 카메라 자세 추정 결과가 즉시 비디오 합성에 사용될 수 있도록 하였다. 제안하는 기법의 유효성 입증을 위해 실사 비디오에 기반한 증강현실시스템을 구현하고 카메라 자세 계산과 비디오 합성의 전체 과정을 단일 프레임에 기반하여 실험을 수행하고 제안 기법의 실용성을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.637-640
• /
• 2022

3차원 물체의 탐지와 자세 추정은 실내외 환경에서 장면 이해, 로봇의 물체 조작 작업, 자율 주행, 증강 현실 등과 같은 다양한 응용 분야들에서 공통적으로 요구되는 매우 중요한 시각 인식 기술이다. 깊이 지도를 요구하는 기존 연구들과는 달리, 본 논문에서는 RGB 컬러 영상만을 이용해 미지의 물체들, 즉 3차원 CAD 모델을 가지고 있지 않은 새로운 물체들을 탐지해내고, 이들의 자세를 추정해낼 수 있는 새로운 신경망 모델을 제안한다. 제안 모델에서는 최근 빠른 속도로 발전하고 있는 깊이 추정 기술을 이용함으로써, 깊이 측정 센서 없이도 물체 자세 추정에 필요한 깊이 지도를 컬러 영상에서 구해낼 수 있다. 본 논문에서는 벤치마크 데이터 집합을 이용한 실험을 통해, 제안 모델의 유용성을 평가한다.

• 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 Convergence for Information Technology
• /
• v.8 no.6
• /
• pp.135-142
• /
• 2018

In recent years, researches utilizing UAV for military purposes such as precision tracking and batting have been actively conducted. In order to track the preceding flight, there has been a previous research on estimating the attitude information of the flight such as roll, pitch, and yaw using images taken from the rear UAV. In this study, we propose a method to estimate the attitude information more precisely by applying the Kalman filter to the existing image processing technique. By applying the Kalman filter to the estimated attitude data using image processing, we could reduce the estimation error of the attitude angle significantly. Through the simulation experiments, it was confirmed that the estimation using the Kalman filter can estimate the posture information of the aircraft more accurately.