본문

서브메뉴

Challenges in 3D Machine Learning: Data Scarcity and Structure-Awareness
Challenges in 3D Machine Learning: Data Scarcity and Structure-Awareness
Challenges in 3D Machine Learning: Data Scarcity and Structure-Awareness

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211153059
ISBN  
9798346390992
DDC  
100
저자명  
Uy, Mikaela Angelina.
서명/저자  
Challenges in 3D Machine Learning: Data Scarcity and Structure-Awareness
발행사항  
[Sl] : Stanford University, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
144 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-05, Section: A.
주기사항  
Advisor: Guibas, Leonidas.
학위논문주기  
Thesis (Ph.D.)--Stanford University, 2024.
초록/해제  
요약3D content creation is an important and increasingly popular task for different applications such as graphic design, gaming and animation, virtual reality experiences, manufacturing, and robotics. This growing popularity creates a strong demand for tools that facilitate the creation and synthesis of new 3D content. Traditionally, designers would need to manually design and create each new model, which is a difficult and labor intensive task, requiring expert knowledge of 3D modeling or computer-aided design (CAD) tools. With the advent of 3D machine learning, there is an opportunity to make the 3D creation process more accessible -- but there are challenges on the way. For example, difficulties arise when 3D data repositories of sufficient quality and size are not available, making approaches that require 3D supervision implausible. Another difficulty occurs when structure-aware, fine-grained control or further manipulation and editing is desired for the output, a task for which coarse homogeneous 3D representations are not ideal. In this dissertation, we address two important challenges in 3D machine learning, namely data scarcity and structure-awareness, thus taking promising steps towards automating more parts of the 3D content creation process.First, we introduce a novel approach for reconstructing a 3D scene when no 3D data is available for direct supervision at training time and instead only a sparse set of image captures are given. We show that by introducing some advances in neural rendering, we are able to coarsely reconstruct a 3D scene from only few 2D input views by modeling ambiguity-aware depth estimates. Moreover, we additionally introduce a new image formation model that allows better neural-rendering-based geometric reconstruction. Second, we address the challenge of creating 3D content that is structure-aware, allowing for the creation of 3D assets with a fine level of geometric detail as well as enabling further user manipulation and editing. Concretely, we propose an approach that leverages existing artist-generated databases (meshes) to create controllable deformations of existing 3D data by conditioning on target images or scans. We also study learning representations of shapes that are more structured, e.g., CAD models, and are directly editable, thus improving user control. Depending on the task, use-case and user preference, these techniques offer different advantages towards making 3D content creation more universally accessible.
일반주제명  
Ambiguity
일반주제명  
Deformation
일반주제명  
Deadlines
일반주제명  
Semantics
일반주제명  
Logic
일반주제명  
Engineering
기타저자  
Stanford University.
기본자료저록  
Dissertations Abstracts International. 86-05A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017164890
■00520250211153059
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798346390992
■035    ▼a(MiAaPQ)AAI31652055
■035    ▼a(MiAaPQ)Stanfordpr895qw3725
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a100
■1001  ▼aUy,  Mikaela  Angelina.
■24510▼aChallenges  in  3D  Machine  Learning:  Data  Scarcity  and  Structure-Awareness
■260    ▼a[Sl]▼bStanford  University▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a144  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-05,  Section:  A.
■500    ▼aAdvisor:  Guibas,  Leonidas.
■5021  ▼aThesis  (Ph.D.)--Stanford  University,  2024.
■520    ▼a3D  content  creation  is  an  important  and  increasingly  popular  task  for  different  applications  such  as  graphic  design,  gaming  and  animation,  virtual  reality  experiences,  manufacturing,  and  robotics.  This  growing  popularity  creates  a  strong  demand  for  tools  that  facilitate  the  creation  and  synthesis  of  new  3D  content.  Traditionally,  designers  would  need  to  manually  design  and  create  each  new  model,  which  is  a  difficult  and  labor  intensive  task,  requiring  expert  knowledge  of  3D  modeling  or  computer-aided  design  (CAD)  tools.  With  the  advent  of  3D  machine  learning,  there  is  an  opportunity  to  make  the  3D  creation  process  more  accessible  --  but  there  are  challenges  on  the  way.  For  example,  difficulties  arise  when  3D  data  repositories  of  sufficient  quality  and  size  are  not  available,  making  approaches  that  require  3D  supervision  implausible.  Another  difficulty  occurs  when  structure-aware,  fine-grained  control  or  further  manipulation  and  editing  is  desired  for  the  output,  a  task  for  which  coarse  homogeneous  3D  representations  are  not  ideal.  In  this  dissertation,  we  address  two  important  challenges  in  3D  machine  learning,  namely  data  scarcity  and  structure-awareness,  thus  taking  promising  steps  towards  automating  more  parts  of  the  3D  content  creation  process.First,  we  introduce  a  novel  approach  for  reconstructing  a  3D  scene  when  no  3D  data  is  available  for  direct  supervision  at  training  time  and  instead  only  a  sparse  set  of  image  captures  are  given.  We  show  that  by  introducing  some  advances  in  neural  rendering,  we  are  able  to  coarsely  reconstruct  a  3D  scene  from  only  few  2D  input  views  by  modeling  ambiguity-aware  depth  estimates.  Moreover,  we  additionally  introduce  a  new  image  formation  model  that  allows  better  neural-rendering-based  geometric  reconstruction.  Second,  we  address  the  challenge  of  creating  3D  content  that  is  structure-aware,  allowing  for  the  creation  of  3D  assets  with  a  fine  level  of  geometric  detail  as  well  as  enabling  further  user  manipulation  and  editing.  Concretely,  we  propose  an  approach  that  leverages  existing  artist-generated  databases  (meshes)  to  create  controllable  deformations  of  existing  3D  data  by  conditioning  on  target  images  or  scans.  We  also  study  learning  representations  of  shapes  that  are  more  structured,  e.g.,  CAD  models,  and  are  directly  editable,  thus  improving  user  control.  Depending  on  the  task,  use-case  and  user  preference,  these  techniques  offer  different  advantages  towards  making  3D  content  creation  more  universally  accessible.
■590    ▼aSchool  code:  0212.
■650  4▼aAmbiguity
■650  4▼aDeformation
■650  4▼aDeadlines
■650  4▼aSemantics
■650  4▼aLogic
■650  4▼aEngineering
■690    ▼a0800
■690    ▼a0395
■690    ▼a0537
■71020▼aStanford  University.
■7730  ▼tDissertations  Abstracts  International▼g86-05A.
■790    ▼a0212
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164890▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF12480 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.