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 101–110 of 111 papers

TitleStatusHype
Towards Difficulty-Agnostic Efficient Transfer Learning for Vision-Language ModelsCode0
What Do You See? Enhancing Zero-Shot Image Classification with Multimodal Large Language ModelsCode0
Language-Driven Anchors for Zero-Shot Adversarial RobustnessCode0
Who's in and who's out? A case study of multimodal CLIP-filtering in DataCompCode0
Open-vocabulary vs. Closed-set: Best Practice for Few-shot Object Detection Considering Text DescribabilityCode0
Multilingual Vision-Language Pre-training for the Remote Sensing DomainCode0
Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training BenchmarkCode0
Text-to-Image Diffusion Models are Zero-Shot ClassifiersCode0
Image-Caption Encoding for Improving Zero-Shot GeneralizationCode0
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP InversionCode0
Show:102550
← PrevPage 11 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