본문

서브메뉴

Discovering the 4D World Behind Any Video
Discovering the 4D World Behind Any Video
Discovering the 4D World Behind Any Video

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20250211152736
ISBN  
9798384448730
DDC  
004
저자명  
Ye, Vickie.
서명/저자  
Discovering the 4D World Behind Any Video
발행사항  
[Sl] : University of California, Berkeley, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
118 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
주기사항  
Advisor: Kanazawa, Angjoo.
학위논문주기  
Thesis (Ph.D.)--University of California, Berkeley, 2024.
초록/해제  
요약As we begin to interact with AI systems, we need them to be able to interpret the visual world in 4D - that is, to perceive the geometry and motion in the world. However, pixel differences in image space result from either geometry (via camera motion) or scene motion in the world. To disentangle these two sources this from a single video is extremely under-constrained.In this thesis, I build several systems that recover scene representations from limited image observations. Specifically, I study a series of problems that build toward the 4D monocular recovery problem, each one addressing a different aspect of the under-constrained nature of the problem. First I study the problem of recovering shape from under-constrained inputs, without scene motion. Specifically, I present pixelNeRF, a method to synthesize novel views of a static scene from single or few views. We learn a scene prior by training a 3D neural representation conditioned on image features across multiple scenes. This learned scene prior enables 3D scene completion from the under-constrained inputs of single or few images. Next I study the problem of recovering motion without 3D shape. In particular, I present Deformable Sprites, a method to extract persistent elements of a dynamic scene from an input video. We represent each element as 2D image layers that deform across the video.Finally I present two studies of performing the joint recovery of both the shape and motion of the 4D world from any single video. I first study the special case of dynamic humans, and present SLAHMR, in which we recover from a single video the global poses of all the humans and the camera in the world coordinate frame. I then move on to the general case of recovering any dynamic objects from a single video in Shape of Motion, in which we recover the entire scene as 4D gaussians, which we can use for dynamic novel view synthesis and 3D tracking.
일반주제명  
Computer science
일반주제명  
Computer engineering
일반주제명  
Information technology
키워드  
3D reconstruction
키워드  
Computer graphics
키워드  
Computer vision
키워드  
Video understanding
키워드  
4D monocular recovery problem
기타저자  
University of California, Berkeley Electrical Engineering & Computer Sciences
기본자료저록  
Dissertations Abstracts International. 86-03B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017163649
■00520250211152736
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384448730
■035    ▼a(MiAaPQ)AAI31491367
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aYe,  Vickie.
■24510▼aDiscovering  the  4D  World  Behind  Any  Video
■260    ▼a[Sl]▼bUniversity  of  California,  Berkeley▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a118  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-03,  Section:  B.
■500    ▼aAdvisor:  Kanazawa,  Angjoo.
■5021  ▼aThesis  (Ph.D.)--University  of  California,  Berkeley,  2024.
■520    ▼aAs  we  begin  to  interact  with  AI  systems,  we  need  them  to  be  able  to  interpret  the  visual  world  in  4D  -  that  is,  to  perceive  the  geometry  and  motion  in  the  world.  However,  pixel  differences  in  image  space  result  from  either  geometry  (via  camera  motion)  or  scene  motion  in  the  world.  To  disentangle  these  two  sources  this  from  a  single  video  is  extremely  under-constrained.In  this  thesis,  I  build  several  systems  that  recover  scene  representations  from  limited  image  observations.  Specifically,  I  study  a  series  of  problems  that  build  toward  the  4D  monocular  recovery  problem,  each  one  addressing  a  different  aspect  of  the  under-constrained  nature  of  the  problem. First  I  study  the  problem  of  recovering  shape  from  under-constrained  inputs,  without  scene  motion.  Specifically,  I  present  pixelNeRF,  a  method  to  synthesize  novel  views  of  a  static  scene  from  single  or  few  views.  We  learn  a  scene  prior  by  training  a  3D  neural  representation  conditioned  on  image  features  across  multiple  scenes.  This  learned  scene  prior  enables  3D  scene  completion  from  the  under-constrained  inputs  of  single  or  few  images.  Next  I  study  the  problem  of  recovering  motion  without  3D  shape.  In  particular,  I  present  Deformable  Sprites,  a  method  to  extract  persistent  elements  of  a  dynamic  scene  from  an  input  video.  We  represent  each  element  as  2D  image  layers  that  deform  across  the  video.Finally  I  present  two  studies  of  performing  the  joint  recovery  of  both  the  shape  and  motion  of  the  4D  world  from  any  single  video.  I  first  study  the  special  case  of  dynamic  humans,  and  present  SLAHMR,  in  which  we  recover  from  a  single  video  the  global  poses  of  all  the  humans  and  the  camera  in  the  world  coordinate  frame.  I  then  move  on  to  the  general  case  of  recovering  any  dynamic  objects  from  a  single  video  in  Shape  of  Motion,  in  which  we  recover  the  entire  scene  as  4D  gaussians,  which  we  can  use  for  dynamic  novel  view  synthesis  and  3D  tracking.
■590    ▼aSchool  code:  0028.
■650  4▼aComputer  science
■650  4▼aComputer  engineering
■650  4▼aInformation  technology
■653    ▼a3D  reconstruction
■653    ▼aComputer  graphics
■653    ▼aComputer  vision
■653    ▼aVideo  understanding
■653    ▼a4D  monocular  recovery  problem
■690    ▼a0984
■690    ▼a0800
■690    ▼a0489
■690    ▼a0464
■71020▼aUniversity  of  California,  Berkeley▼bElectrical  Engineering  &  Computer  Sciences.
■7730  ▼tDissertations  Abstracts  International▼g86-03B.
■790    ▼a0028
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163649▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Подробнее информация.

    • Бронирование
    • не существует
    • моя папка
    • Первый запрос зрения
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    материал
    Reg No. Количество платежных Местоположение статус Ленд информации
    TF09703 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Бронирование доступны в заимствований книги. Чтобы сделать предварительный заказ, пожалуйста, нажмите кнопку бронирование

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.