• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.8
• /
• pp.5695-5706
• /
• 2015

Recently adoption of BIM technology for building renovation and remodeling has been increased in construction industry. However most buildings have trouble in 2D drawing-based BIM modeling, because 2D drawings have not been updated real situations continually. Applying reverse engineering, this study analysed the point cloud data structure and the process for guideline of reverse engineering of architectural MEP, and deducted the relating considerations. To active usage of 3D scanning technique in domestic, the objective of this study is to analyze the point cloud data processing from real site with terrestrial LiDAR and the process from data gathering to data acquisition.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.125-127
• /
• 2021

본 논문에서는 기존의 산업 및 서비스 변화에 따라 발생하는 빅데이터 분석 서비스 처리를 위한 빅데이터 분석 서비스 브로커 시스템을 제안한다. 기존의 빅데이터 분석 시스템은 분석하는 시간 동안 지속적으로 자원을 점유하고 있어야 하며, 이러한 서비스를 이용하기 위해 내부에 대규모의 시스템을 구축하고 지속적으로 운영해야하는 단점이 존재한다. 본 논문에서는 빅데이터 분석에 필요한 자원을 효과적으로 사용하기 위해 클라우드 기반의 자원 관리와 연계하고 서비스 이용을 용이하게 하기 위해 단일 엔드포인트 기반의 빅데이터 분석 서비스 호출 구조를 설계하였다. 이를 통해 빅데이터 서비스 분석에 소요되는 자원 점유에 따라 동적으로 자원을 생성 관리하여 자원을 보다 효과적으로 이용할 수 있는지 테스트베드를 구축하여 서비스 이용 및 자원 사용을 효과적으로 하는지 확인하였다. 또한, 이를 통해 대규모 자원을 지속적으로 점유해야하는 빅데이터 분석 플랫폼의 자원사용에 대한 한계를 일부 해소하여 자원을 효과적으로 이용할 수 있는 것을 확인하였다.

• Journal of Platform Technology
• /
• v.8 no.4
• /
• pp.11-19
• /
• 2020

In an autonomous driving environment, dynamic recognition of objects is essential as the situation changes in real time. In addition, as the number of sensors and control modules built into an autonomous vehicle increases, the amount of data the central control unit has to process also rapidly increases. By minimizing the output data from the sensor, the load on the central control unit can be reduced. This study proposes a dynamic object recognition algorithm solely using the embedded processor on a LiDAR sensor. While there are open source algorithms to process the point cloud output from LiDAR sensors, most require a separate high-performance processor. Since the embedded processors installed in LiDAR sensors often have resource constraints, it is essential to optimize the algorithm for efficiency. In this study, an embedded processor based object recognition algorithm was developed for autonomous vehicles, and the correlation between the size of the point clouds and processing time was analyzed. The proposed object recognition algorithm evaluated that the processing time directly increased with the size of the point cloud, with the processor stalling at a specific point if the point cloud size is beyond the threshold

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.345-352
• /
• 2022

Recent LiDAR(Light Detection And Ranging) sensor is used for scanning object around in real-time. This sensor can detect movement of the object and how it has changed. As the production cost of the sensors has been decreased, LiDAR begins to be used for various industries such as facility guard, smart city and self-driving car. However, LiDAR has a large input data size due to its real-time scanning process. So another way for processing a large amount of data are needed in LiDAR system because it can cause a bottleneck. This paper proposes edge computing to compress massive point cloud for processing quickly. Since laser's reflection range of LiDAR sensor is limited, multiple LiDAR should be used to scan a large area. In this reason multiple LiDAR sensor's data should be processed at once to detect or recognize object in real-time. Edge computer compress point cloud efficiently to accelerate data processing and decompress every data in the main cloud in real-time. In this way user can control LiDAR sensor in the main system without any bottleneck. The system we suggest solves the bottleneck which was problem on the cloud based method by applying edge computing service.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.486-493
• /
• 2023

In this paper, we propose an automatic hand gesture area extraction and recognition technique using FMCW radar-based point cloud and LSTM. The proposed technique has the following originality compared to existing methods. First, unlike methods that use 2D images as input vectors such as existing range-dopplers, point cloud input vectors in the form of time series are intuitive input data that can recognize movement over time that occurs in front of the radar in the form of a coordinate system. Second, because the size of the input vector is small, the deep learning model used for recognition can also be designed lightly. The implementation process of the proposed technique is as follows. Using the distance, speed, and angle information measured by the FMCW radar, a point cloud containing x, y, z coordinate format and Doppler velocity information is utilized. For the gesture area, the hand gesture area is automatically extracted by identifying the start and end points of the gesture using the Doppler point obtained through speed information. The point cloud in the form of a time series corresponding to the viewpoint of the extracted gesture area is ultimately used for learning and recognition of the LSTM deep learning model used in this paper. To evaluate the objective reliability of the proposed technique, an experiment calculating MAE with other deep learning models and an experiment calculating recognition rate with existing techniques were performed and compared. As a result of the experiment, the MAE value of the time series point cloud input vector + LSTM deep learning model was calculated to be 0.262 and the recognition rate was 97.5%. The lower the MAE and the higher the recognition rate, the better the results, proving the efficiency of the technique proposed in this paper.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.7
• /
• pp.335-347
• /
• 2024

Open-vocabulary 3D point cloud instance segmentation (OV-3DIS) is a challenging visual task to segment a 3D scene point cloud into object instances of both base and novel classes. In this paper, we propose a novel model Open3DME for OV-3DIS to address important design issues and overcome limitations of the existing approaches. First, in order to improve the quality of class-agnostic 3D masks, our model makes use of T3DIS, an advanced Transformer-based 3D point cloud instance segmentation model, as mask proposal module. Second, in order to obtain semantically text-aligned visual features of each point cloud segment, our model extracts both 2D and 3D features from the point cloud and the corresponding multi-view RGB images by using pretrained CLIP and OpenSeg encoders respectively. Last, to effectively make use of both 2D and 3D visual features of each point cloud segment during label assignment, our model adopts a unique feature ensemble method. To validate our model, we conducted both quantitative and qualitative experiments on ScanNet-V2 benchmark dataset, demonstrating significant performance gains.

• Broadcasting and Media Magazine
• /
• v.26 no.2
• /
• pp.46-55
• /
• 2021

최근 디바이스와 5G 통신의 비약적인 발전을 통해 가상/증강 현실 분야, 자율 주행 등 3차원 그래픽스 기술에 대한 연구가 활발하게 진행되고 있으며, 3차원 정보를 면밀하게 표현할 수 있는 포인트 클라우드와 다시점 초실감 콘텐츠가 주목받고 있다. 이와 같은 콘텐츠는 전통적인 2D 비디오 대비 많은 데이터를 사용하고 있기에, 효율적 사용을 위해서는 압축이 필수적으로 요구된다. 이에 따라 국제표준화기구인 ISO/IEC 산하 Moving Picture Expert Group(MPEG)에서는 고밀도 포인트 클라우드 및 초다시점 실감형 콘텐츠에 대한 압축 방안으로 V-PCC(Video based Point Cloud Compression) 및 MIV(MPEG Immersive Video) 기술을 표준화 중에 있으며, 또한, 압축된 데이터를 효율적으로 저장, 전송하기 위한 방안으로 Carriage of Visual Volumetric Video Coding(V3C) 표준화가 진행중에 있다. 본 고에서는 MPEG에서 진행중인 V3C 표준 기술에 대하여 살펴보고자 한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.323-324
• /
• 2023

본 논문에서는 다양한 복원 인공지능 알고리즘 중 하나인 3차원 복원 기술은 실제로 존재하는 물체의 2차원적인 픽셀을 3차원의 형태로 구현하여 형상화한다. 정확한 3차원 정보 처리가 요구됨에 따라 포인트 클라우드로 표현되는 데이터를 통해 정확한 쿨체의 크기 정보나 좌표 정보를 표시할 수 있다. 데이터의 픽셀을 분석하여 3차원의 형태로 구현할 것을 정의하는 복셀화(Voxelization) 알고리즘 전처리 과정을 통해 3차원 복원 기술 3D-GAN 활용으로 3차원 형태 형상화를 하였다. 본 논문에서는 3차원 복원 알고리즘 통하여 2차원 포인트 클라우드를 분석해 3차원 형태로 복원하는 기술에 대한 설명한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.5
• /
• pp.1199-1206
• /
• 2015

This paper introduces an algorithm for computing similarity between two poses in the motion capture data with different scale of skeleton, different number of joints and different joint names. The proposed algorithm first performs the topological analysis on the skeleton hierarchy for classifying the joints into more meaningful groups. The global joints positions of each joint group then are aggregated into a point cloud. The number of joints and their positions are automatically adjusted in this process. Once we have two point clouds, the algorithm finds an optimal 2D transform matrix that transforms one point cloud to the other as closely as possible. Then, the similarity can be obtained by summing up all distance values between two points clouds after applying the 2D transform matrix. After some experiment, we found that the proposed algorithm is able to compute the similarity between two poses regardless of their scale, joint name and the number of joints.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.6
• /
• pp.571-581
• /
• 2020

Vegetation affects water level change and flow resistance in rivers and impacts waterway ecosystems as a whole. Therefore, it is important to have accurate information about the species, shape, and size of any river vegetation. However, it is not easy to collect full vegetation data on-site, so recent studies have attempted to obtain large amounts of vegetation data using terrestrial laser scanning (TLS). Also, due to the complex shape of vegetation, it is not easy to obtain accurate information about the canopy area, and there are limitations due to a complex range of variables. Therefore, the physical structure of vegetation was analyzed in this study by reconfiguring high-resolution point cloud data collected through 3-dimensional terrestrial laser scanning (3D TLS) in a voxel. Each physical structure was analyzed under three different conditions: a simple vegetation formation without leaves, a complete formation with leaves, and a patch-scale vegetation formation. In the raw data, the outlier and unnecessary data were filtered and removed by Statistical Outlier Removal (SOR), resulting in 17%, 26%, and 25% of data being removed, respectively. Also, vegetation volume by voxel size was reconfigured from post-processed point clouds and compared with vegetation volume; the analysis showed that the margin of error was 8%, 25%, and 63% for each condition, respectively. The larger the size of the target sample, the larger the error. The vegetation surface looked visually similar when resizing the voxel; however, the volume of the entire vegetation was susceptible to error.