Neidio i’r brif dudalen lywio Neidio i chwilio Neidio i’r prif gynnwys

Knowledge Distillation Classifier Generation Network for Zero-Shot Learning

  • Yunlong Yu
  • , Bin Li
  • , Zhong Ji
  • , Jungong Han
  • , Zhongfei Zhang
  • Zhejiang University
  • Tianjin University
  • Binghamton University

Allbwn ymchwil: Cyfraniad at gyfnodolyn › Erthygl › adolygiad gan gymheiriaid

17 Dyfyniadau (Scopus)
96 Wedi eu Llwytho i Lawr (Pure)

Crynodeb

In this article, we present a conceptually simple but effective framework called knowledge distillation classifier generation network (KDCGN) for zero-shot learning (ZSL), where the learning agent requires recognizing unseen classes that have no visual data for training. Different from the existing generative approaches that synthesize visual features for unseen classifiers' learning, the proposed framework directly generates classifiers for unseen classes conditioned on the corresponding class-level semantics. To ensure the generated classifiers to be discriminative to the visual features, we borrow the knowledge distillation idea to both supervise the classifier generation and distill the knowledge with, respectively, the visual classifiers and soft targets trained from a traditional classification network. Under this framework, we develop two, respectively, strategies, i.e., class augmentation and semantics guidance, to facilitate the supervision process from the perspectives of improving visual classifiers. Specifically, the class augmentation strategy incorporates some additional categories to train the visual classifiers, which regularizes the visual classifier weights to be compact, under supervision of which the generated classifiers will be more discriminative. The semantics-guidance strategy encodes the class semantics into the visual classifiers, which would facilitate the supervision process by minimizing the differences between the generated and the real-visual classifiers. To evaluate the effectiveness of the proposed framework, we have conducted extensive experiments on five datasets in image classification, i.e., AwA1, AwA2, CUB, FLO, and APY. Experimental results show that the proposed approach performs best in the traditional ZSL task and achieves a significant performance improvement on four out of the five datasets in the generalized ZSL task.

Iaith wreiddiolSaesneg
Tudalennau (o-i)3183-3194
Nifer y tudalennau12
CyfnodolynIEEE Transactions on Neural Networks and Learning Systems
Cyfrol34
Rhif cyhoeddi6
Dyddiad ar-lein cynnar29 Medi 2021
Dynodwyr Gwrthrych Digidol (DOIs)
StatwsCyhoeddwyd - 01 Meh 2023

Ôl bys

Gweld gwybodaeth am bynciau ymchwil 'Knowledge Distillation Classifier Generation Network for Zero-Shot Learning'. Gyda’i gilydd, maen nhw’n ffurfio ôl bys unigryw.

Dyfynnu hyn