SOTAVerified

Zero-Shot Image Classification

Zero-shot image classification is a technique in computer vision where a model can classify images into categories that were not present during training. This is achieved by leveraging semantic information about the categories, such as textual descriptions or relationships between classes.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 41–50 of 111 papers

TitleStatusHype
Can We Talk Models Into Seeing the World Differently?Code1
Interpreting and Analysing CLIP's Zero-Shot Image Classification via Mutual KnowledgeCode1
LiT: Zero-Shot Transfer with Locked-image text TuningCode1
PerceptionCLIP: Visual Classification by Inferring and Conditioning on ContextsCode1
TaxaBind: A Unified Embedding Space for Ecological ApplicationsCode1
A Progressive Framework of Vision-language Knowledge Distillation and Alignment for Multilingual Scene—0
Language to Network: Conditional Parameter Adaptation with Natural Language Descriptions—0
Large-Scale Zero-Shot Image Classification from Rich and Diverse Textual Descriptions—0
DiRaC-I: Identifying Diverse and Rare Training Classes for Zero-Shot Learning—0
CSA: Data-efficient Mapping of Unimodal Features to Multimodal Features—0
Show:102550
← PrevPage 5 of 12Next →
All datasetsCountry211ICinWODinW

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1OpenClip H/14 (34B)(Laion2B)Top-1 accuracy30.01—Unverified
#ModelMetricClaimedVerifiedStatus
1CLIP (ViT B-32)Average Score56.64—Unverified
#ModelMetricClaimedVerifiedStatus
1GLIP (Tiny A)Average Score11.4—Unverified