권오빈 교수
Oh-Bin Kwon
서울대학교 생화학과 · 의학
연구실 소개
권오빈 교수의 연구실은 신경생물학과 인공지능 기반 시각적 인식 기술을 융합한 연구를 주도하고 있습니다. 정신질환의 신경생물학적 기반을 밝히는 데 초점을 맞춘 신경전달물질 및 수용체 기반 메커니즘 연구와 동시에, 시각 기반 자율주행 및 환경 인식을 위한 지도 기반 레이더 지도 및 그래프 기반 메모리 구조 설계 등 첨단 시각 인식 기술 개발도 함께 진행하고 있습니다. 특히, 뇌 기능과 인공지능의 상호보완적 접근을 통해 인지 기능 향상과 지능형 로봇 주행 기술의 혁신을 추구하고 있습니다.
연구 현황
연구 성과 추이
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
주요 논문
15Energy expenditure and energy intake need to be balanced to maintain proper energy homeostasis. Energy homeostasis is tightly regulated by the central nervous system, and the hypothalamus is the primary center for the regulation of energy balance. The hypothalamus exerts its effect through both humoral and neuronal mechanisms, and each hypothalamic area has a distinct role in the regulation of energy expenditure. Recent studies have advanced the understanding of the molecular regulation of energ
We present a novel graph-structured memory for visual navigation, called visual graph memory (VGM), which consists of unsupervised image representations obtained from navigation history. The proposed VGM is constructed incrementally based on the similarities among the unsupervised representations of observed images, and these representations are learned from an unlabeled image dataset. We also propose a navigation framework that can utilize the proposed VGM to tackle visual navigation problems.
CONTEXT: It was previously reported in Korea that there were 1.4 case per million per year of acromegaly. This was low in comparison with the extrapolated values of Western European countries. We expected that the incidence of acromegaly would be much higher now because of recently improved medical facilities, diagnostic tools and coverage of medical insurance to all the population of South Korea. OBJECTIVE: The purpose of this nationwide survey was to examine the incidence and prevalence of pat
We propose a novel type of map for visual navigation, a renderable neural radiance map (RNR-Map), which is designed to contain the overall visual information of a 3D environment. The RNR-Map has a grid form and consists of latent codes at each pixel. These latent codes are embedded from image observations, and can be converted to the neural radiance field which enables image rendering given a camera pose. The recorded latent codes implicitly contain visual information about the environment, whic
CONTEXT: Serotonin acts as a neurotransmitter in the central and enteric nervous systems, modulating psychological, metabolic and gastrointestinal functions. Serotonin is also found in the serum or plasma, indicating its potential role as a hormone. OBJECTIVE: We aimed to assess the 24-hour diurnal profile of serum serotonin in relation to meal ingestion in healthy adult men. METHODS: Ten healthy (5 lean and 5 obese) male subjects were enrolled in this study. Blood samples were drawn every 30-60
Both osteoporosis and depression are major health threats, but their interrelationship is not clear. This study elucidated the associations between osteoporosis and depression while considering the temporal sequence of the diagnoses. In this cross-sectional study, data were extracted from the Korean National Health and Nutrition Examination Surveys (2007-2009 and 2015-2019, n = 29,045). Osteoporosis and depression were defined by diagnoses thereof. The odds ratio (OR) of the incident osteoporosi
The dipeptidyl peptidase-4 (DPP4) inhibitors have become widely used antidiabetic medication. They control glycemia by interacting with serum DPP4 to interfere catalyzation of incretins. The aim of this pilot study was to discover the DPP4 polymorphisms that could affect the efficacy of vildagliptin, a DPP-4 inhibitor in diabetic patients.
BACKGROUND: Age-related primary sarcopenia and end-stage renal disease (ESRD)-related muscle wasting are discrete entities; however, both manifest as a decline in skeletal muscle mass and strength. The etiological pathways differ, with aging factors implicated in sarcopenia and a combination of uremic factors, including haemodialysis, contributing to ESRD-related muscle wasting. Understanding these molecular nuances is imperative for targeted interventions, and the integration of proteomic and m
This paper tackles a localization problem in large-scale indoor environments with wayfinding maps. A wayfinding map abstractly portrays the environment, and humans can localize themselves based on the map. However, when it comes to using it for robot localization, large geometrical discrepancies between the wayfinding map and the real world make it hard to use conventional localization methods. Our objective is to estimate a robot pose within a wayfinding map, utilizing RGB images from perspecti
대표 연구 분야
권오빈 교수의 연구를 Nubint에서 더 깊이 살펴보세요
이 연구실의 논문을 앱에서 열어 AI와 함께 읽고, 핵심을 요약하고, 내 글에 인용하세요.