DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 안성진 | - |
dc.contributor.author | Kim, Yeongbin | - |
dc.contributor.author | 김영빈 | - |
dc.date.accessioned | 2024-07-30T19:31:44Z | - |
dc.date.available | 2024-07-30T19:31:44Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1097253&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/321673 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전산학부, 2024.2,[iv, 40 p. :] | - |
dc.description.abstract | Systematic compositionality, or the ability to adapt to novel situations by creating a mental model of the world using reusable pieces of knowledge, remains a significant challenge in machine learning. While there has been considerable progress in the language domain, efforts towards systematic visual imagination, or envisioning the dynamical implications of a visual observation, are in their infancy. We introduce the Systematic Visual Imagination Benchmark (SVIB), the first benchmark designed to address this problem head-on. SVIB offers a novel framework for a minimal world modeling problem, where models are evaluated based on their ability to generate one-step image-to-image transformations under a latent world dynamics. The framework provides benefits such as the possibility to jointly optimize for systematic perception and imagination, a range of difficulty levels, and the ability to control the fraction of possible factor combinations used during training. We provide a comprehensive evaluation of various baseline models on SVIB, offering insight into the current state-of-the-art in systematic visual imagination. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 체계적 구성성▼a체계적 일반화▼a벤치마크▼a월드 모델링 | - |
dc.subject | Systematic compositionality▼aVisual imagination▼aBenchmark▼aWorld modeling | - |
dc.title | (A) study on visual systematic generalization via one-step image generation world model | - |
dc.title.alternative | 이미지 생성 월드모델을 통한 시각정보의 체계적 일반화 연구 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | Ahn, Sungjin | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.