최소 단어 이상 선택하여야 합니다.
최대 10 단어까지만 선택 가능합니다.
다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
NTIS 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
DataON 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Edison 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Kafe 바로가기주관연구기관 | 고려대학교 Korea University |
---|---|
연구책임자 | 도락주 |
보고서유형 | 최종보고서 |
발행국가 | 대한민국 |
언어 | 한국어 |
발행년월 | 2019-08 |
과제시작연도 | 2019 |
주관부처 | 과학기술정보통신부 Ministry of Science and ICT |
등록번호 | TRKO202000001189 |
과제고유번호 | 1711082654 |
사업명 | 글로벌프론티어지원(R&D) |
DB 구축일자 | 2020-05-30 |
키워드 | 위치인식.환경 모델링.객체지향.물체인식.구조 인식.SLAM.3D Scanner.RGB-Depth.3D realistic map.localization.environmental modeling.object-oriented.object recognition.structure recognition. |
Geometric Primitive로 구성 가능한 다중 가림 실내 환경에서 고정밀-강인성-확장성을 확보한 Immersive 3D Mapping 기술 상용화 완료.
1. 지도제작 알고리즘 및 viewer 고도화를 통한 정밀성 향상
2. 대형 환경에서도 안정적 데이터 취득 및 처리 가능
3. 지도제작 알고리즘의 표준연계 및 글로벌 기업과의 기술 교류
(출처 : 보고서 요약서 3p)
□ Purpose & Contents
This research aims at developing a mobile 3D indoor scanner for realistic environmental modeling in the object-oriented way and an automatic mapping method using the sensor data obtained from the scanner. The goal of this research is that the 3D mesh of a RGB-Depth map is aut
□ Purpose & Contents
This research aims at developing a mobile 3D indoor scanner for realistic environmental modeling in the object-oriented way and an automatic mapping method using the sensor data obtained from the scanner. The goal of this research is that the 3D mesh of a RGB-Depth map is automatically created when an mobile 3D indoor scanner obtains the data statically by going here and there in any vista-scale indoor living space with the portable scanner.
The contents of this research are composed of the following four parts.
The first is to develop mobile 3D scanner which contains following advantages.
1) Expansion of user‘s degree of Freedom through vertically data acquisition.
2) Reduction data acquisition time through remote control system.
3) Minimization data storage.
Though these advantages, we could collect realistic data more quickly. Using these data and World best SLAM technology developed from our research team, we could built World best accurate map.
The second is building a neat structure map like a papered house. After extraction of the structure of indoor environment, a texture-mapped mesh model of the structure is created. A neat mesh model is made by hole filling based on reasoning occluded parts.
The third is object SCAN(Segmentation-Classification-Alternation-Notching). It is a process that each individual object divided from multiple objects is classified and the substituted DB model of it is disposed. Through the process, the map of the above environment is automatically built with realistic mesh of objects.
Last is to experience Modeling environment using HMD. To do this process, we combined our conventional algorithms and implemented series of processes. Under restricted environment, we verified similarity of depth and color between modeling environment and real environment.
□ Results
Our research team will research above 4 parts to achieve project purpose. Using these Development results, our research team expect following results.
Trough research Developments, we expect academic outcomes such as submit a SCI journals, present a paper at the international or domestic conference. Also apply for a international or domestic patent on its Development results, we could expect technology transfer.
□ Expected Contribution
Automatic 3D realistic RGB-Depth mapping is possible through the above technology. The map can be used for realistic contents of HMD, 3D map service, remote environment sharing and localization using a smart phone. Also new services based on BIM can be created in architectural field like making or updating CAD, inspection of completion, etc. It can be applied to a base technology of DIY(Do It Yourself) market that a consumer builds a 3D map personally and shares it.
(출처 : SUMMARY 5p)
과제명(ProjectTitle) : | - |
---|---|
연구책임자(Manager) : | - |
과제기간(DetailSeriesProject) : | - |
총연구비 (DetailSeriesProject) : | - |
키워드(keyword) : | - |
과제수행기간(LeadAgency) : | - |
연구목표(Goal) : | - |
연구내용(Abstract) : | - |
기대효과(Effect) : | - |
Copyright KISTI. All Rights Reserved.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.