Show simple item record

dc.contributor.authorElhoseiny, Mohamed
dc.contributor.authorYi, Kai
dc.contributor.authorElfeki, Mohamed
dc.date.accessioned2021-01-06T08:48:52Z
dc.date.available2021-01-06T08:48:52Z
dc.date.issued2021-01-01
dc.identifier.citationhttps://openaccess.thecvf.com/content_ICCV_2019/papers/Elhoseiny_Creativity_Inspired_Zero-Shot_Learning_ICCV_2019_paper.pdf
dc.identifier.urihttp://hdl.handle.net/10754/666828
dc.description.abstractZero-shot learning (ZSL) aims at understanding unseen categories with no training examples from class-level descriptions. To improve the discriminative power of ZSL, we model the visual learning process of unseen categories with inspiration from the psychology of human creativity for producing novel art. First, we propose CIZSL-v1 as a creativity inspired model for generative ZSL. We relate ZSL to human creativity by observing that ZSL is about recognizing the unseen, and creativity is about creating a likable unseen. We introduce a learning signal inspired by creativity literature that explores the unseen space with hallucinated class-descriptions and encourages careful deviation of their visual feature generations from seen classes while allowing knowledge transfer from seen to unseen classes. Second, CIZSL-v2 is proposed as an improved version of CIZSL-v1 for generative zero-shot learning. CIZSL-v2 consists of an investigation of additional inductive losses for unseen classes along with a semantic guided discriminator. Empirically, we show consistently that CIZSL losses can improve generative ZSL models on the challenging task of generalized ZSL from a noisy text on CUB and NABirds datasets. We also show the advantage of our approach to Attribute-based ZSL on AwA2, aPY, and SUN datasets. We also show that CIZSL-v2 has improved performance compared to CIZSL-v1.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/2101.00173
dc.rightsArchived with thanks to arXiv
dc.titleCIZSL++: Creativity Inspired Generative Zero-Shot Learning
dc.typePreprint
dc.contributor.departmentComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
dc.contributor.departmentKing Abdullah University Of Science And Technology , KAUST, Thuwal, Saudi Arabia.
dc.eprint.versionPre-print
dc.contributor.institutionComputer Science department, The University of central Florida.
dc.identifier.arxivid2101.00173
kaust.personElhoseiny, Mohamed
kaust.personYi, Kai
refterms.dateFOA2021-01-06T08:49:38Z


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
3.810Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record