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 81–90 of 111 papers

TitleStatusHype
I2DFormer: Learning Image to Document Attention for Zero-Shot Image Classification—0
PaLI: A Jointly-Scaled Multilingual Language-Image ModelCode0
What does a platypus look like? Generating customized prompts for zero-shot image classificationCode2
Zero-Shot Temporal Action Detection via Vision-Language PromptingCode1
DUET: Cross-modal Semantic Grounding for Contrastive Zero-shot LearningCode1
Disentangled Ontology Embedding for Zero-shot LearningCode1
Masked Unsupervised Self-training for Label-free Image ClassificationCode1
CCMB: A Large-scale Chinese Cross-modal BenchmarkCode1
PyramidCLIP: Hierarchical Feature Alignment for Vision-language Model Pretraining—0
Zero-Shot Logit AdjustmentCode1
Show:102550
← PrevPage 9 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