SOTAVerified

Image Classification

Image Classification is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike object detection, which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as image retrieval, which also involves finding similar images in a large database.

Source: Metamorphic Testing for Object Detection Systems

Papers

Showing 226250 of 10419 papers

TitleStatusHype
Contrastive Learning Rivals Masked Image Modeling in Fine-tuning via Feature DistillationCode2
Matryoshka Representation LearningCode2
Fast Vision Transformers with HiLo AttentionCode2
Inception TransformerCode2
ConvMAE: Masked Convolution Meets Masked AutoencodersCode2
Masked Generative DistillationCode2
Deep PCB To COCO ConvertorCode2
CLIP-Art: Contrastive Pre-training for Fine-Grained Art ClassificationCode2
Understanding The Robustness in Vision TransformersCode2
K-LITE: Learning Transferable Visual Models with External KnowledgeCode2
Neighborhood Attention TransformerCode2
Masked Siamese Networks for Label-Efficient LearningCode2
DaViT: Dual Attention Vision TransformersCode2
Solving ImageNet: a Unified Scheme for Training any Backbone to Top ResultsCode2
Unified Contrastive Learning in Image-Text-Label SpaceCode2
Rethinking Visual Geo-localization for Large-Scale ApplicationsCode2
BatchFormerV2: Exploring Sample Relationships for Dense Representation LearningCode2
MultiMAE: Multi-modal Multi-task Masked AutoencodersCode2
Contrastive learning of Class-agnostic Activation Map for Weakly Supervised Object Localization and Semantic SegmentationCode2
Focal Modulation NetworksCode2
Decoupled Knowledge DistillationCode2
Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNsCode2
QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training QuantizationCode2
Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timeCode2
ParC-Net: Position Aware Circular Convolution with Merits from ConvNets and TransformerCode2
Show:102550
← PrevPage 10 of 417Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1CoCa (finetuned)Top 1 Accuracy91Unverified
2Model soups (BASIC-L)Top 1 Accuracy90.98Unverified
3Model soups (ViT-G/14)Top 1 Accuracy90.94Unverified
4DaViT-GTop 1 Accuracy90.4Unverified
5Meta Pseudo Labels (EfficientNet-L2)Top 1 Accuracy90.2Unverified
6DaViT-HTop 1 Accuracy90.2Unverified
7SwinV2-GTop 1 Accuracy90.17Unverified
8MAWS (ViT-6.5B)Top 1 Accuracy90.1Unverified
9Florence-CoSwin-HTop 1 Accuracy90.05Unverified
10RevCol-HTop 1 Accuracy90Unverified