메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술대회자료
저자정보
Jinhoon Wang (Kwangwoon University) Woonghee Lee (Kwangwoon University) Made Putra Arya Winata (Kwangwoon University) Junghyun Oh (Kwangwoon University)
저널정보
제어로봇시스템학회 제어로봇시스템학회 국제학술대회 논문집 ICCAS 2024
발행연도
2024.10
수록면
1,552 - 1,555 (4page)

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
Active 3D reconstruction is a research field that focuses on planning paths to actively acquire data efficiently for 3D reconstruction. Among these, Next-Best-View (NBV) based path planning studies methods to select best viewpoints for 3D reconstruction. Recently, realistic modeling techniques such as Neural Radiance Fields and Gaussian Splatting have emerged, leading to active research on applying NBV to new 3D modeling methods. However, recent NBV techniques are designed to consider only uncertainty when selecting the best image from candidate images, which limits their efficiency in data acquisition. To overcome this limitation, we propose a similar image avoidance policy using SSIM, which avoids selecting images similar to those used in training. This approach prevents overfitting and effectively reduces the number of data used in training while acquiring the overall shape and surface information of the model with a small amount of data. We then verify the performance of the proposed method by comparing it to 3D models generated with random data and FisherRF using image quality metrics.

목차

Abstract
1. INTRODUCTION
2. METHOD
3. EXPERIMENT
4. CONCLUSION
REFERENCES

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

최근 본 자료

전체보기

댓글(0)

0