• 한국CDE학회지
• /
• 제16권2호
• /
• pp.2-7
• /
• 2010

• 방송공학회논문지
• /
• 제26권6호
• /
• pp.692-703
• /
• 2021

최근 컴퓨터 그래픽 기술의 발전으로 현실의 객체를 더욱 사실적인 가상의 그래픽으로 표현하는 기술의 연구가 활발히 진행되고 있다. 포인트 클라우드는 3차원 공간 좌표와 색 정보 등을 포함하는 수많은 점을 사용해 3차원 객체를 표현하는 기술로 기존의 2차원 영상보다 많은 데이터를 사용하고 데이터 처리에 더욱 복잡한 연산이 필요하므로 포인트 클라우드를 사용한 서비스를 제공하기 위해서는 거대한 데이터 저장 공간과 높은 성능의 연산 장치가 필요하다. 현재 국제 표준기구인 MPEG에서 포인트 클라우드를 2차원 평면에 투영한 다음 비디오 코덱을 사용해 압축하는 Video-based Point Cloud Compression (V-PCC) 기술이 연구되고 있다. V-PCC 기술은 포인트 클라우드를 점유 맵 (Occupancy map), 기하 영상 (Geometry image), 속성 영상 (Attribute image) 등의 2차원 영상과 2차원 영상과 3차원 공간 사이의 관계를 알려주는 보조 정보를 사용해 압축한다. 복호화된 포인트 클라우드의 밀도를 높이거나 객체를 확대할 때, 일반적으로 3차원 연산을 사용하지만 연산 방식이 복잡하고 많은 시간을 소모하며 새로운 포인트의 정확한 생성 위치를 결정하기 힘들다는 한계가 존재한다. 이에 본 논문은 V-PCC의 포인트 클라우드가 투영된 영상에 2차원 보간 (Interpolation) 기술을 적용해 적은 연산으로 보다 정확한 추가 포인트 클라우드를 생성하는 방안을 제안한다.

• 로봇학회논문지
• /
• 제17권3호
• /
• pp.373-380
• /
• 2022

This paper considers a method of fast correspondence matching for iterative closest point (ICP) algorithm. In robotics, the ICP algorithm and its variants have been widely used for pose estimation by finding the translation and rotation that best align two point clouds. In computational perspectives, the main difficulty is to find the correspondence point on the reference point cloud to each observed point. Jump-table-based correspondence matching is one of the methods for reducing computation time. This paper proposes a method that corrects errors in an existing jump-table-based correspondence matching algorithm. The criterion activating the use of jump-table is modified so that the correspondence matching can be applied to the situations, such as point-cloud registration problems with highly curved surfaces, for which the existing correspondence-matching method is non-applicable. For demonstration, both hardware and simulation experiments are performed. In a hardware experiment using Hokuyo-10LX LiDAR sensor, our new algorithm shows 100% correspondence matching accuracy and 88% decrease in computation time. Using the F1TENTH simulator, the proposed algorithm is tested for an autonomous driving scenario with 2D range-bearing point cloud data and also shows 100% correspondence matching accuracy.

• 방송공학회논문지
• /
• 제25권1호
• /
• pp.25-35
• /
• 2020

최근 하드웨어 성능뿐 아니라 영상 처리 기술의 발달로 인해 다양한 분야에서 사용자에게 자유로운 시야각과 입체감을 제공하는 3차원 포인트를 다루는 기술에 관한 연구를 지속하고 있다. 3차원 포인트를 표현하는 형식 중 포인트 클라우드 기술은 포인트를 정밀하게 획득/표현할 수 있다는 장점으로 인해 다양한 분야에서 주목받고 있다. 하지만 하나의 3차원 포인트 클라우드 콘텐츠를 표현하기 위해 수십, 수백만 개의 포인트가 필요하므로 기존의 2차원 콘텐츠보다 많은 양의 저장 공간을 요구한다는 단점이 존재한다. 이러한 이유로, 국제 표준화 기구인 MPEG (Moving Picture Experts Group)에서는 3차원 포인트 클라우드 콘텐츠를 효율적으로 압축 및 저장하고, 사용자에게 전송하는 방안에 대해 계속 연구를 진행 중이다. 본 논문에서는 MPEG-I (Immersive) 그룹에서 제안한 V-PCC(Video based Point Cloud Compression) 부호화기를 통해 생성된 V-PCC 비트스트림을 MMT (MPEG Media Transport) 표준에서 정의한 MPU (Media Processing Unit)로 구성하는 방안을 제안한다. 또한, MMT 표준에서 정의한 시그널링 메시지를 확장하여 3차원 포인트 클라우드 콘텐츠의 영역 선별적 전송 방안을 위한 파라미터와 사용자의 요구에 따라 선택적으로 품질 파라미터를 결정할 수 있도록 V-PCC에서 상정하는 품질 파라미터를 추가 정의한다. 마지막으로, 본 논문에서는 제안한 기술을 기반으로 검증 플랫폼의 설계/구현을 통해 결과를 확인한다.

• Journal of Information Processing Systems
• /
• 제13권3호
• /
• pp.491-499
• /
• 2017

In this study, we proposed a new approach to segment ground and nonground points gained from a 3D laser range sensor. The primary aim of this research was to provide a fast and effective method for ground segmentation. In each frame, we divide the point cloud into small groups. All threshold points and start-ground points in each group are then analyzed. To determine threshold points we depend on three features: gradient, lost threshold points, and abnormalities in the distance between the sensor and a particular threshold point. After a threshold point is determined, a start-ground point is then identified by considering the height difference between two consecutive points. All points from a start-ground point to the next threshold point are ground points. Other points are nonground. This process is then repeated until all points are labelled.

• 대한의용생체공학회:의공학회지
• /
• 제44권5호
• /
• pp.346-353
• /
• 2023

In this paper, we propose an innovative approach that leverages deep learning to find optimal reference points for achieving precise tooth segmentation in three-dimensional tooth point cloud data. A dataset consisting of 350 aligned maxillary and mandibular cloud data was used as input, and both end coordinates of individual teeth were used as correct answers. A two-dimensional image was created by projecting the rendered point cloud data along the Z-axis, where an image of individual teeth was created using an object detection algorithm. The proposed algorithm is designed by adding various modules to the Unet model that allow effective learning of a narrow range, and detects both end points of the tooth using the generated tooth image. In the evaluation using DSC, Euclid distance, and MAE as indicators, we achieved superior performance compared to other Unet-based models. In future research, we will develop an algorithm to find the reference point of the point cloud by back-projecting the reference point detected in the image in three dimensions, and based on this, we will develop an algorithm to divide the teeth individually in the point cloud through image processing techniques.

• International journal of advanced smart convergence
• /
• 제9권3호
• /
• pp.192-198
• /
• 2020

The 3D point cloud is a key technology of object detection for virtual reality and augmented reality. In order to apply various areas of object detection, it is necessary to obtain 3D information and even color information more easily. In general, to generate a 3D point cloud, it is acquired using an expensive scanner device. However, 3D and characteristic information such as RGB and depth can be easily obtained in a mobile device. GNN (Graph Neural Network) can be used for object detection based on these characteristics. In this paper, we have generated RGB and RGBD by detecting basic information and characteristic information from the KITTI dataset, which is often used in 3D point cloud object detection. We have generated RGB-GNN with i-GNN, which is the most widely used LiDAR characteristic information, and color information characteristics that can be obtained from mobile devices. We compared and analyzed object detection accuracy using RGBD-GNN, which characterizes color and depth information.

• 한국정밀공학회지
• /
• 제23권9호
• /
• pp.200-207
• /
• 2006

In this paper a new approach which combines implicit surface scheme and NURBS surface interpolation method is proposed in order to generate a complete surface model from unorganized point cloud data. In the method a base surface was generated by creating smooth implicit surface from the input point cloud data through which the actual surface would pass. The implicit surface was defined by a combination of shape functions including quadratic polynomial function, cubic polynomial functions and radial basis function using adaptive domain decomposition method. In this paper voxel data which can be extracted easily from the base implicit surface were used in order to generate rectangular net with good quality using the normal projection and smoothing scheme. After generating the interior points and tangential vectors in each rectangular region considering the required accuracy, the NURBS surface were constructed by interpolating the rectangular array of points using boundary tangential vectors which assure C$^1$ continuity between rectangular patches. The validity and effectiveness of this new approach was demonstrated by performing numerical experiments for the various types of point cloud data.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.422-424
• /
• 2021

In this paper, it leads the approach of fusing multiple RGB cameras for visual objects recognition based on deep learning with convolution neural network and 3D Light Detection and Ranging (LiDAR) to observe the environment and match into a 3D world in estimating the distance and position in a form of point cloud map. The goal of perception in multiple cameras are to extract the crucial static and dynamic objects around the autonomous vehicle, especially the blind spot which assists the AV to navigate according to the goal. Numerous cameras with object detection might tend slow-going the computer process in real-time. The computer vision convolution neural network algorithm to use for eradicating this problem use must suitable also to the capacity of the hardware. The localization of classified detected objects comes from the bases of a 3D point cloud environment. But first, the LiDAR point cloud data undergo parsing, and the used algorithm is based on the 3D Euclidean clustering method which gives an accurate on localizing the objects. We evaluated the method using our dataset that comes from VLP-16 and multiple cameras and the results show the completion of the method and multi-sensor fusion strategy.

• 국제학술발표논문집
• /
• The 8th International Conference on Construction Engineering and Project Management
• /
• pp.3-12
• /
• 2020

Nowadays, as-is BIM generation has been popularly adopted in the architecture, engineering, construction and facility management (AEC/FM) industries. In order to generate a 3D as-is BIM of a structural component, current methods require a registration process that merges different sets of point cloud data obtained from multiple locations, which is time-consuming and registration error-prone. To tackle this limitation, this study proposes a registration-free 3D point cloud data acquisition technique for as-is BIM generation. In this study, small-size mirrors that rotate in both horizontal and vertical direction are used to enable the registration-free data acquisition technique. First, a geometric model that defines the relationship among the mirrors, the laser scanner and the target component is developed. Second, determinations of optimal laser scanner location and mirror location are performed based on the developed geometrical model. To validate the proposed registration-free as-is BIM generation technique, simulation tests are conducted on key construction components including a PC slab and a structural wall. The result demonstrates that the registration-free point cloud data acquisition technique can be applicable in various construction elements including PC elements and structural components for as-is BIM generation.