본문

서브메뉴

Language Supervision for Computer Vision
Language Supervision for Computer Vision
Language Supervision for Computer Vision

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211152057
ISBN  
9798382739380
DDC  
004
저자명  
Desai, Karan P.
서명/저자  
Language Supervision for Computer Vision
발행사항  
[Sl] : University of Michigan, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
179 p
주기사항  
Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
주기사항  
Advisor: Johnson, Justin C.
학위논문주기  
Thesis (Ph.D.)--University of Michigan, 2024.
초록/해제  
요약Representation learning lies at the core of modern Artificial Intelligence. In computer vision, labeled image datasets like ImageNet have been the standard choice for representation learning. Despite being empirically successful, this approach is expensive to scale due to labeling costs. Moreover, the representation quality is limited by the size and diversity of datasets and their associated label ontologies.My research explores using natural language supervision for computer vision. Using natural language allows us to go beyond fixed label ontologies and scale up to more general sources such as internet data. Toward this goal, my dissertation explores four problems - (1) Learning representations: I propose one of the first methods for language-supervised visual learning that uses image captioning as the training objective, showing its efficacy compared to ImageNet-trained methods on downstream tasks like object detection and segmentation. (2) Scaling data: I explore social media as a rich source of high-quality image descriptions and curate a dataset of 12 million image-text pairs while ensuring responsible curation practices. (3) Understanding data: It is difficult to comprehend the diversity of visual concepts present in millions of image-text pairs. I posit that images and text naturally organize into a tree-like hierarchy and propose an approach for learning representations that capture this hierarchy using tools from hyperbolic geometry. (4) Transfer to downstream tasks: Large vision-language models show impressive zero-shot transfer capabilities on image-level tasks like classification and retrieval. However, their transferability to pixel-level tasks like object detection and segmentation has relied on expensive labeled mask annotations. I propose an object detector to efficiently transfer pre-trained vision models to segment and classify visual objects without any fine-tuning, unlike existing detectors that train using orders of magnitude more labeled masks to achieve high performance.In summary, my research affirms that using language supervision can drive the next leap of progress in computer vision and has immense utility in practical applications.
일반주제명  
Computer science
일반주제명  
Computer engineering
키워드  
Computer vision
키워드  
Representation learning
키워드  
Hyperbolic geometry
키워드  
Natural language
기타저자  
University of Michigan Computer Science & Engineering
기본자료저록  
Dissertations Abstracts International. 85-12B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017162812
■00520250211152057
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798382739380
■035    ▼a(MiAaPQ)AAI31348980
■035    ▼a(MiAaPQ)umichrackham005447
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aDesai,  Karan  P.
■24510▼aLanguage  Supervision  for  Computer  Vision
■260    ▼a[Sl]▼bUniversity  of  Michigan▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a179  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  85-12,  Section:  B.
■500    ▼aAdvisor:  Johnson,  Justin  C.
■5021  ▼aThesis  (Ph.D.)--University  of  Michigan,  2024.
■520    ▼aRepresentation  learning  lies  at  the  core  of  modern  Artificial  Intelligence.  In  computer  vision,  labeled  image  datasets  like  ImageNet  have  been  the  standard  choice  for  representation  learning.  Despite  being  empirically  successful,  this  approach  is  expensive  to  scale  due  to  labeling  costs.  Moreover,  the  representation  quality  is  limited  by  the  size  and  diversity  of  datasets  and  their  associated  label  ontologies.My  research  explores  using  natural  language  supervision  for  computer  vision.  Using  natural  language  allows  us  to  go  beyond  fixed  label  ontologies  and  scale  up  to  more  general  sources  such  as  internet  data.  Toward  this  goal,  my  dissertation  explores  four  problems  -  (1)  Learning  representations:  I  propose  one  of  the  first  methods  for  language-supervised  visual  learning  that  uses  image  captioning  as  the  training  objective,  showing  its  efficacy  compared  to  ImageNet-trained  methods  on  downstream  tasks  like  object  detection  and  segmentation.  (2)  Scaling  data:  I  explore  social  media  as  a  rich  source  of  high-quality  image  descriptions  and  curate  a  dataset  of  12  million  image-text  pairs  while  ensuring  responsible  curation  practices.  (3)  Understanding  data:  It  is  difficult  to  comprehend  the  diversity  of  visual  concepts  present  in  millions  of  image-text  pairs.  I  posit  that  images  and  text  naturally  organize  into  a  tree-like  hierarchy  and  propose  an  approach  for  learning  representations  that  capture  this  hierarchy  using  tools  from  hyperbolic  geometry.  (4)  Transfer  to  downstream  tasks:  Large  vision-language  models  show  impressive  zero-shot  transfer  capabilities  on  image-level  tasks  like  classification  and  retrieval.  However,  their  transferability  to  pixel-level  tasks  like  object  detection  and  segmentation  has  relied  on  expensive  labeled  mask  annotations.  I  propose  an  object  detector  to  efficiently  transfer  pre-trained  vision  models  to  segment  and  classify  visual  objects  without  any  fine-tuning,  unlike  existing  detectors  that  train  using  orders  of  magnitude  more  labeled  masks  to  achieve  high  performance.In  summary,  my  research  affirms  that  using  language  supervision  can  drive  the  next  leap  of  progress  in  computer  vision  and  has  immense  utility  in  practical  applications.
■590    ▼aSchool  code:  0127.
■650  4▼aComputer  science
■650  4▼aComputer  engineering
■653    ▼aComputer  vision
■653    ▼aRepresentation  learning
■653    ▼aHyperbolic  geometry
■653    ▼aNatural  language
■690    ▼a0984
■690    ▼a0800
■690    ▼a0464
■71020▼aUniversity  of  Michigan▼bComputer  Science  &  Engineering.
■7730  ▼tDissertations  Abstracts  International▼g85-12B.
■790    ▼a0127
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162812▼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
    TF10452 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.