- 실생활에 로봇을 사용하기 위해서는 사람과 물체를 인식하고, 공간 구조를 파악 및 판단하는 다양한 인지 기술이 필요합니다. PEARL은 다양한 센서를 활용한 로봇 비전 연구를 수행하여 로봇을 인간처럼 똑똑하게 만드는 것을 목표로 합니다.
지도교수 : 현장훈
연구결과
대규모 실내 환경에서의 카메라 기반 위치추정 기술 개발 (Benchmark 챌린지 1위 달성)
VR/AR를 위한 3차원 실감 지도 생성 기술 개발
Smart band를 활용한 수면 패턴 분석 기술 개발
이미지 및 3차원 데이터의 동적 객체 추출 기술 개발
논문
Hyewon Han, Dongsin Kim, JinSeok Kim, Lee Ku Kwac, Janghun Hyeon, Junhyoung Oh, “A novel sleep aid device to reduce sleep latency using air-CO2 mixed gas,” Frontiers in Neurology,May 2023
Janghun Hyeon, Bumchul Jang, Hyunga Choi, Joohyung Kim, Dongwoo Kim, Nakju Doh, “Photo-realistic 3D model based accurate visual positioning system for large-scale indoor spaces,” Engineering Applications of Artificial Intelligence,Apr 2023.
Woodo Lee, Hyejin Kim, Jaekwoun Shim, Dongsin Kim, Janghun Hyeon, Eunyeon Joo, Byung-Euk Joo, Junhyoung Oh, “The simplification of the insomnia severity index and Epworth sleepiness scale using machine learning models,” Scientific Reports, Apr 2023.
Joohyung Kim, Janghun Hyeon, Hyunga Choi, Bumchul Jang, Bokyeon Jeong, Nakju Doh, “Cloc: confident initial estimation of long-term visual localization using a few sequential images in large-scale spaces,” IEEE Sensors Journal, Mar 2023.
Janghun Hyeon, Joohyung Kim, Hyunga Choi, Bumchul Jang, Jaehyeon Kang, Nakju Doh, “Spatial template-based geometric complexity reduction method for photo-realistic modeling of large-scale indoor spaces,” Engineering Applications of Artificial Intelligence,Nov 2022.
Woonhyung Jung, Janghun Hyeon, Nakju Doh, “Robust Cuboid Modeling from Noisy and Incomplete 3D point clouds using Gaussian mixture model,” Remote Sensing,Oct 2022
Janghun Hyeon, Joohyung Kim, Nakju Doh, “Pose correction for highly accurate visual localization in large-scale indoor spaces,” IEEE/CVF International Conference on Computer Vision (ICCV), 2021.
Joohyung Kim, Janghun Hyeon, Nakju Doh, “Generative multiviewinpaintingfor object removal in large indoor spaces,” International Journal of Advanced Robotic System,Mar 2021.
Janghun Hyeon, Dongwoo Kim, Bumchul Jang, Hyunga Choi, Dong Hoon Yi, Kyungho Yoo, Jeongae Choi, Nakju Doh, “KR-Net: A Dependable Visual Kidnap Recovery Network for Indoor Spaces,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020.
Janghun Hyeon, Hyunga Choi, Joohyung Kim, Bumchul Jang, Jaehyeon Kang, Nakju Doh, “Automatic spatial template generation for realistic 3D modeling of large-scale indoor spaces,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019
Janghun Hyeon, Weonsuk Lee, Joohyung Kim, Nakju Doh, “NormNet: Point-wise normal estimation network for three-dimensional point cloud data,” International Journal of Advanced Robotic Systems, Jul 2019.
Keonyong Lee, Soo-Hyun Ryu, Suyong Yeon, HyunGi Cho, ChangHyun Jun, Jaehyeon Kang, Hyunga Choi, Janghun Hyeon, Insik Beak, Woonhyung Jung, Hanul Kim, Nakju Doh, “Accurate continuous sweeping framework in indoor spaces with backpack sensor system for applications to 3D mapping,” IEEE Robotics and Automation Letters, Jan 2016.
특허
Method and apparatus for estimating user pose using three-dimensional virtual space model, 2022.09.08. (Japan)
3차원 가상 공간 모델을 이용한 사용자 포즈 추정 방법 및 장치, 2022.04.13. (Korea)
3차원 지도작성을 위한 데이터 수집 장치 2019.03.21. (Korea)
기술이전
Camera-Lidar calibration algorithm to generate precise RGB-D point clouds (2017)
3D map generation and device for indoor spaces(2019)