본문

서브메뉴

Learning Visual Concepts
Learning Visual Concepts
Learning Visual Concepts

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20250211152037
ISBN  
9798384023104
DDC  
004
저자명  
Shivakumar, Shreyas Skandan.
서명/저자  
Learning Visual Concepts
발행사항  
[Sl] : University of Pennsylvania, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
198 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-02, Section: A.
주기사항  
Advisor: Taylor, Camillo J.
학위논문주기  
Thesis (Ph.D.)--University of Pennsylvania, 2024.
초록/해제  
요약We propose a framework to use off-the-shelf pre-trained object detection models and extend them for use on unseen datasets in a manner requiring little to no modification of the original architecture, and by adding only a few additional components to the overall pipeline. Motivated by the role of attributes in zero-shot-learning paradigms, we define conceptual groups by using positive and negative exemplars retroactively, and evaluate the feasibility of recognizing a variety of these proposed conceptual groups in a corpus of previously unseen data, including unseen categories. We conduct experiments with networks trained on the COCO dataset, and utilize Open-Images-V7 as our held out unseen dataset. Our analysis suggests that existing off-the-shelf object detection networks such as Faster-RCNN can be leveraged to extract useful information beyond the scope of a straightforward category prediction framework. This information can be used to operationalize the idea of concept learning through a set of positive and negative exemplars and a simple linear SVM operating on the features produced by the deep network. We compare this approach to vision enabled large language models such as LLaVA, CogVLM and GPT4V, and show a strong baseline performance with lower resource requirements. Additionally, we illustrate that this method can be scaled to larger concept sets by validating this approach on a larger set of concepts in the LVIS dataset. We illustrate a few approaches to better understand the semantic topology of their learned feature space, and we measure the feasibility of using these features for the identification of the proposed conceptual groups. We propose strategies to leverage this information to predict these conceptual groups on previously unseen samples containing unseen class categories.
일반주제명  
Computer science
일반주제명  
Robotics
일반주제명  
Information science
키워드  
Classification
키워드  
Computer vision
키워드  
Machine perception
키워드  
Object detection
키워드  
Object recognition
기타저자  
University of Pennsylvania Computer and Information Science
기본자료저록  
Dissertations Abstracts International. 86-02A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017162643
■00520250211152037
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384023104
■035    ▼a(MiAaPQ)AAI31335541
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aShivakumar,  Shreyas  Skandan.
■24510▼aLearning  Visual  Concepts
■260    ▼a[Sl]▼bUniversity  of  Pennsylvania▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a198  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-02,  Section:  A.
■500    ▼aAdvisor:  Taylor,  Camillo  J.
■5021  ▼aThesis  (Ph.D.)--University  of  Pennsylvania,  2024.
■520    ▼aWe  propose  a  framework  to  use  off-the-shelf  pre-trained  object  detection  models  and  extend  them  for  use  on  unseen  datasets  in  a  manner  requiring  little  to  no  modification  of  the  original  architecture,  and  by  adding  only  a  few  additional  components  to  the  overall  pipeline.  Motivated  by  the  role  of  attributes  in  zero-shot-learning  paradigms,  we  define  conceptual  groups  by  using  positive  and  negative  exemplars  retroactively,  and  evaluate  the  feasibility  of  recognizing  a  variety  of  these  proposed  conceptual  groups  in  a  corpus  of  previously  unseen  data,  including  unseen  categories.  We  conduct  experiments  with  networks  trained  on  the  COCO  dataset,  and  utilize  Open-Images-V7  as  our  held  out  unseen  dataset.  Our  analysis  suggests  that  existing  off-the-shelf  object  detection  networks  such  as  Faster-RCNN  can  be  leveraged  to  extract  useful  information  beyond  the  scope  of  a  straightforward  category  prediction  framework.  This  information  can  be  used  to  operationalize  the  idea  of  concept  learning  through  a  set  of  positive  and  negative  exemplars  and  a  simple  linear  SVM  operating  on  the  features  produced  by  the  deep  network.  We  compare  this  approach  to  vision  enabled  large  language  models  such  as  LLaVA,  CogVLM  and  GPT4V,  and  show  a  strong  baseline  performance  with  lower  resource  requirements.  Additionally,  we  illustrate  that  this  method  can  be  scaled  to  larger  concept  sets  by  validating  this  approach  on  a  larger  set  of  concepts  in  the  LVIS  dataset.  We  illustrate  a  few  approaches  to  better  understand  the  semantic  topology  of  their  learned  feature  space,  and  we  measure  the  feasibility  of  using  these  features  for  the  identification  of  the  proposed  conceptual  groups.  We  propose  strategies  to  leverage  this  information  to  predict  these  conceptual  groups  on  previously  unseen  samples  containing  unseen  class  categories.
■590    ▼aSchool  code:  0175.
■650  4▼aComputer  science
■650  4▼aRobotics
■650  4▼aInformation  science
■653    ▼aClassification
■653    ▼aComputer  vision
■653    ▼aMachine  perception
■653    ▼aObject  detection
■653    ▼aObject  recognition
■690    ▼a0800
■690    ▼a0984
■690    ▼a0771
■690    ▼a0723
■71020▼aUniversity  of  Pennsylvania▼bComputer  and  Information  Science.
■7730  ▼tDissertations  Abstracts  International▼g86-02A.
■790    ▼a0175
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162643▼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
    TF09946 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.