• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.481-486
• /
• 2007

본 논문에서는 깊이 영상을 개선하는 방법으로 깊이 영상 획득 시 손실된 영역을 복원하는 기법을 제안한다. 대상 객체의 동적인 3차원 정보는 적외선 깊이 센서가 장착된 깊이 비디오 카메라를 통하여 실시간으로 획득한다. 이때, 깊이 비디오뿐만 아니라 각 프레임마다 컬러영상이 동시에 획득된다. 그러나 대상 객체의 일부 또는 전체가 반짝이는 검은 재질로 되어있을 경우, 획득된 깊이 영상에 손실이 발생한다. 특히 방송용 콘텐츠로서 연기자의 3차원 정보를 획득할 때 머리카락 영역이 손실되는 심각한 문제가 발생한다. 이를 해결하기 위해 먼저 컬러 영상을 이용하여 손실된 영역의 위치 정보를 알아낸다. 손실된 영역 내 경계부분의 깊이 정보를 복원한 후 2차 베지어 커브로 보간하여 내부의 깊이 정보를 복원한다. 개선된 깊이 영상을 기반으로 일련의 모델링 과정을 수행하면 보다 자연스러운 3차원 모델을 생성할 수 있다. 생성된 3차원 모델은 실감방송용 콘텐츠로 사용될 수 있으며, 시청자에게 시각상호작용과 촉각상호작용 등 다차원 감각의 상호작용을 제공할 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.733-735
• /
• 2004

이 논문은 한 대의 카메라에서 얻은 일련의 영상을 해석하여 단순한 2차원의 영상을 3차원물체로 복원하는 방법에 대해 설명을 한다. 이러한 3차원 복원 방법은 카메라 내부 변수가 동일하다는 가정을 이용하여 별도의 캘리브레이션 작업 없이 한 대의 카메라로부터 얻은 여러 장의 영상을 이용한다. 이 논문에서 제안한 방법은 내부 변수 중 카메라 행렬의 단순화와 사영 기하를 이용한 것이다 이 방법은 실제 비디오 프레임에 가상의 그래픽 모델을 더하는 AR (Augmented reality) 분야에 특히 유용하다 이 논문에서의 실험은 실제 여러 비디오 스트림 데이터를 바탕으로 수행되었고, 하나의 카메라를 사용한 동영상에서 3차원 구조로 복원하는 실험 결과는 시스템의 유용성을 보여준다.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.264-272
• /
• 2019

In this paper, we propose a 3D human body reconstruction and refinement method from the frames extracted from a video to obtain natural and smooth motion in temporal domain. Individual frames extracted from the video are fed into convolutional neural network to estimate the location of the joint and the silhouette of the human body. This is done by projecting the parameter-based 3D deformable model to 2D image and by estimating the value of the optimal parameters. If the reconstruction process for each frame is performed independently, temporal consistency of human pose and shape cannot be guaranteed, yielding an inaccurate result. To alleviate this problem, the proposed method analyzes and interpolates the principal component parameters of the 3D morphable model reconstructed from each individual frame. Experimental result shows that the erroneous frames are corrected and refined by utilizing the relation between the previous and the next frames to obtain the improved 3D human reconstruction result.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.150-161
• /
• 2013

This paper presents a new system which produces high resolution 3D contents by capturing multiview images of an object using multiple cameras, and estimating geometric and texture information of the object from the captured images. Even though a variety of multiview image-based 3D reconstruction systems have been proposed, it was difficult to generate high resolution 3D contents because multiview image-based 3D reconstruction requires a large amount of memory and computation. In order to reduce computational complexity and memory size for 3D reconstruction, the proposed system predetermines the regions in input images where an object can exist to extract object boundaries fast. And for fast computation of a visual hull, the system represents silhouettes and 3D-2D projection/back-projection relations by chain codes and 1D homographies, respectively. The geometric data of the reconstructed object is compactly represented by a 3D segment-based data format which is called DoCube, and the 3D object is finally reconstructed after 3D mesh generation and texture mapping are performed. Experimental results show that the proposed system produces 3D object contents of $800{\times}800{\times}800$ resolution with a rate of 2.2 seconds per frame.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.122-131
• /
• 2011

A widely used single image-based 3D face reconstruction method, 3D morphable shape model, reconstructs an accurate 3D facial shape when 2D facial feature points are correctly extracted from an input face image. However, in the case that a user's cooperation is not available such as a real-time 3D face reconstruction system, this method can be vulnerable to the errors of automatic facial feature point extraction. In order to solve this problem, we automatically classify extracted facial feature points into two groups, erroneous and correct ones, and then reconstruct a 3D facial shape by using only the correctly extracted facial feature points. The experimental results showed that the 3D reconstruction performance of the proposed method was remarkably improved compared to that of the previous method which does not consider the errors of automatic facial feature point extraction.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.07a
• /
• pp.417-418
• /
• 2012

본 논문에서는 3차원 데이터를 이용한 효율적인 실내 공간 표현 기법을 제안한다. 제안하는 기법은 3차원 데이터의 획득과 실내 구조 및 영상 정보를 표현하기 위한 표현 복원으로 구성된다. 3차원 데이터는 레이저 거리 측정기(laser range finder, LRF)와 전방향(omni) 카메라를 통해 획득한 포인트 클라우드 공간 정보와 전방향 텍스쳐 영상으로 구성된다. 실내 구조를 복원하기 위해, 획득한 포인트 클라우드를 복셀 격자 기반의 샘플링 기법을 통해 균일화하고 포아송 표면 재구성(Poisson surface rocoostruction) 기법을 통해 3차원 메쉬를 생성한다. 그리고 전방향 텍스쳐 영상과 3차원 메쉬외 기하학적 관계를 이용한 텍스쳐 매핑 기법을 통해 최종적으로 3차원 메쉬 표면을 복원한다. 실험 결과를 통해 제안하는 기법이 실내 공간을 효과적으로 표현함을 확인한다.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.6
• /
• pp.26-34
• /
• 2008

The information of traversability and path planning is essential for UGV(Unmanned Ground Vehicle) navigation. Such information can be obtained by analyzing 3D terrain. In this paper, we present the method of 3D terrain modeling with color information from a camera, precise distance information from a 2D Laser Range Finder(LRF) and wheel encoder information from mobile robot with less data. And also we present the method of 3B terrain modeling with the information from GPS/IMU and 2D LRF with less data. To fuse the color information from camera and distance information from 2D LRF, we obtain extrinsic parameters between a camera and LRF using planar pattern. We set up such a fused system on a mobile robot and make an experiment on indoor environment. And we make an experiment on outdoor environment to reconstruction 3D terrain with 2D LRF and GPS/IMU(Inertial Measurement Unit). The obtained 3D terrain model is based on points and requires large amount of data. To reduce the amount of data, we use cubic grid-based model instead of point-based model.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.3
• /
• pp.27-33
• /
• 2003

The image-based 3D modeling is the technique of generating a 3D graphic model from images acquired using cameras. It is being researched as an alternative technique for the expensive 3D scanner. In this paper, we propose the image-based, 3D modeling system using calibrated stereo cameras. The proposed algorithm for rendering, 3D model consists of three steps, camera calibration, 3D reconstruction, and 3D registration step. In the camera calibration step, we estimate the camera matrix for the image aquisition camera. In the 3D reconstruction step, we calculate 3D coordinates using triangulation from corresponding points of the stereo image. In the 3D registration step, we estimate the transformation matrix that transforms individually reconstructed 3D coordinates to the reference coordinate to render the single 3D model. As shown the result, we generated relatively accurate 3D model.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.11
• /
• pp.47-55
• /
• 2012

In this paper, we propose a method to reconstruct accurate 3D shape object by using multi-view images which are disturbed by motion blur. In multi-view deblurring, more precise PSF estimation can be done by using the geometric relationship between multi-view images. The proposed method first estimates initial 2D PSFs from individual input images. Then 3D PSF candidates are projected on the input images one by one to find the best one which are mostly consistent with the initial 2D PSFs. 3D PSF consists with direction and density and it represents the 3D trajectory of object motion. 야to restore 3D shape by using multi-view images computes the similarity map and estimates the position of 3D point. The estimated 3D PSF is again projected to input images and they replaces the intial 2D PSFs which are finally used in image deblurring. Experimental result shows that the quality of image deblurring and 3D reconstruction improves significantly compared with the result when the input images are independently deblurred.

• Journal of the Institute of Convergence Signal Processing
• /
• v.6 no.3
• /
• pp.112-119
• /
• 2005

In this paper, a model of neural circuit was proposed, which abstracts the depth information in two images gotten from right and left retinas. The proposed neural circuit corresponds to binocular stereo vision based on psychologic and physiological knowledge, and we examine a restoration method of three-dimensional surface. In case of drawing a disparity based on characteristics of images, we can not abstract the depth information correctly if resemblant characteristics are repeated on the boundary region of an object. A binocular disparity is decided in a model of neural circuit by abstraction, synthesis, and correction of a disparity. And we propose a method which restores three-dimensional shape by correcting a depth information, and also restores a three-dimensional surface by mapping a left input image on the restored three-dimensional shape. And we confirmed that the computation time for disparity abstraction can be greatly reduced through the simulation.