Using weighted ranking with classification for facial age estimation가중 랭킹과 분류 방법을 활용한 얼굴 나이 인식 방법

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 466
  • Download : 0
DC FieldValueLanguage
dc.contributor.advisorYang, Hyun Seung-
dc.contributor.advisor양현승-
dc.contributor.authorIm, Woobin-
dc.date.accessioned2019-09-04T02:47:04Z-
dc.date.available2019-09-04T02:47:04Z-
dc.date.issued2018-
dc.identifier.urihttp://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=734093&flag=dissertationen_US
dc.identifier.urihttp://hdl.handle.net/10203/267061-
dc.description학위논문(석사) - 한국과학기술원 : 전산학부, 2018.2,[ii, 19 p. :]-
dc.description.abstractWe propose an end-to-end deep learning framework for age estimation using face images. Our key observation is that ranking face images by age plays an important role for learning features and estimating age. We thus exploit a ranking objective jointly with an age classification objective. In this joint configuration, the ranking objective provides relative information to a deep model, that produces higher accuracy. For the ranking objective, we use a triplet ranking strategy with two novel schemes: relative triplet selection and weighted triplet ranking loss. First, the relative triplet selection expands a pool of possible triplets, enabling effective learning for ranking. Second, the weighted triplet ranking loss reflects the relativeness of age and considers its varying importance for learning. We have applied our method to two famous age estimation benchmarks, Adience and FG-NET, and demonstrated that our approach achieves meaningful improvement over the state-of-the-art age estimation methods.-
dc.languageeng-
dc.publisher한국과학기술원-
dc.subjectFacial age recognition▼adeep learning▼aranking loss▼aartificial neural network-
dc.subject얼굴 나이 인식▼a딥 러닝▼a랭킹 손실 함수▼a인공 신경망-
dc.titleUsing weighted ranking with classification for facial age estimation-
dc.title.alternative가중 랭킹과 분류 방법을 활용한 얼굴 나이 인식 방법-
dc.typeThesis(Master)-
dc.identifier.CNRN325007-
dc.description.department한국과학기술원 :전산학부,-
dc.contributor.alternativeauthor임우빈-
Appears in Collection
CS-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0