DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Chang-Ick | - |
dc.contributor.advisor | 김창익 | - |
dc.contributor.author | Ahn, Il-Koo | - |
dc.contributor.author | 안일구 | - |
dc.date.accessioned | 2015-04-23T06:12:55Z | - |
dc.date.available | 2015-04-23T06:12:55Z | - |
dc.date.issued | 2014 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=591815&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/196586 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 전기및전자공학과, 2014.8, [ xi, 83 p. ] | - |
dc.description.abstract | Free-viewpoint rendering (FVR) has become a popular topic in 3D research. A promising technology in FVR is to generate virtual views using a single texture image and the corresponding depth image. A critical problem that occurs when generating virtual views is that the regions covered by the foreground objects in the original view may be disoccluded in the synthesized views. In this paper, a depth based disocclusion filling algorithm using patch based texture synthesis is proposed. In contrast to the existing patch based virtual view synthesis methods, the filling priority is driven by the robust structure tensor which efficiently reflects the overall structure of an image part and a new confidence term which produces fine synthesis results even near the foreground boundaries. Moreover, the best-matched patch is searched in the background regions and finally it is chosen through a new patch distance measure. Significant superiority of the proposed method over the state-of-the-art methods is presented by comparing the experimental results. We further provide the utility of the proposed method by applying it to 3D video call system with a single camera. | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Disocclusion Filling | - |
dc.subject | 얼굴 추출 | - |
dc.subject | 다중 분할 | - |
dc.subject | 헤시안 행렬 | - |
dc.subject | 인페인팅 | - |
dc.subject | 가려짐 영역 채움법 | - |
dc.subject | Inpainting | - |
dc.subject | Hessian Matrix | - |
dc.subject | Multiple Segmentations | - |
dc.subject | Face Segmentation | - |
dc.title | Depth-aided disocclusion filling using exemplar-based inpainting improved by hessian matrix | - |
dc.title.alternative | 헤시안 행렬을 통해 개선된 예제기반 인페인팅을 이용한 깊이영상 보조의 가려짐 영역 채움법 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 591815/325007 | - |
dc.description.department | 한국과학기술원 : 전기및전자공학과, | - |
dc.identifier.uid | 020085387 | - |
dc.contributor.localauthor | Kim, Chang-Ick | - |
dc.contributor.localauthor | 김창익 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.