SOTAVerified

Semi-Supervised Image Classification

Semi-supervised image classification leverages unlabelled data as well as labelled data to increase classification performance.

You may want to read some blog posts to get an overview before reading the papers and checking the leaderboards:

( Image credit: Self-Supervised Semi-Supervised Learning )

Papers

Showing 150 of 167 papers

TitleStatusHype
ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse LabelsCode0
Applications and Effect Evaluation of Generative Adversarial Networks in Semi-Supervised Learning0
Simple Semi-supervised Knowledge Distillation from Vision-Language Models via Dual-Head OptimizationCode0
Weakly Semi-supervised Whole Slide Image Classification by Two-level Cross Consistency Supervision0
Diff-SySC: An Approach Using Diffusion Models for Semi-Supervised Image Classification0
SynCo: Synthetic Hard Negatives in Contrastive Learning for Better Unsupervised Visual RepresentationsCode0
Self Adaptive Threshold Pseudo-labeling and Unreliable Sample Contrastive Loss for Semi-supervised Image Classification0
A Method of Moments Embedding Constraint and its Application to Semi-Supervised LearningCode0
InfoMatch: Entropy Neural Estimation for Semi-Supervised Image ClassificationCode1
Pseudo-label Learning with Calibrated Confidence Using an Energy-based Model0
Color-S^4L: Self-supervised Semi-supervised Learning with Image Colorization0
Roll With the Punches: Expansion and Shrinkage of Soft Label Selection for Semi-supervised Fine-Grained LearningCode1
Meta Co-Training: Two Views are Better than OneCode1
SequenceMatch: Revisiting the design of weak-strong augmentations for Semi-supervised learningCode0
Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble ProjectorCode0
SemiReward: A General Reward Model for Semi-supervised LearningCode1
Towards Semi-supervised Learning with Non-random Missing LabelsCode1
How To Overcome Confirmation Bias in Semi-Supervised Image Classification By Active Learning0
SimMatchV2: Semi-Supervised Learning with Graph ConsistencyCode1
Shrinking Class Space for Enhanced Certainty in Semi-Supervised LearningCode1
NP-SemiSeg: When Neural Processes meet Semi-Supervised Semantic SegmentationCode1
Scaling Up Semi-supervised Learning with Unconstrained Unlabelled DataCode0
RelationMatch: Matching In-batch Relationships for Semi-supervised LearningCode0
Graph Convolutional Networks based on Manifold Learning for Semi-Supervised Image Classification0
VNE: An Effective Method for Improving Deep Representation by Manipulating Eigenvalue DistributionCode1
NP-Match: Towards a New Probabilistic Model for Semi-Supervised LearningCode1
Learning Customized Visual Models with Retrieval-Augmented KnowledgeCode1
Semi-MAE: Masked Autoencoders for Semi-supervised Vision Transformers0
Self Meta Pseudo Labels: Meta Pseudo Labels Without The Teacher0
Beyond ADMM: A Unified Client-variance-reduced Adaptive Federated Learning Framework0
SVFormer: Semi-supervised Video Transformer for Action RecognitionCode1
Semi-Supervised Single-View 3D Reconstruction via Prototype Shape PriorsCode1
OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation LearningCode0
USB: A Unified Semi-supervised Learning Benchmark for ClassificationCode3
Semi-supervised Vision Transformers at ScaleCode1
RDA: Reciprocal Distribution Alignment for Robust Semi-supervised LearningCode0
Semi-Supervised Hyperspectral Image Classification Using a Probabilistic Pseudo-Label Generation FrameworkCode1
NP-Match: When Neural Processes meet Semi-Supervised LearningCode1
FreeMatch: Self-adaptive Thresholding for Semi-supervised LearningCode3
DoubleMatch: Improving Semi-Supervised Learning with Self-SupervisionCode0
Masked Siamese Networks for Label-Efficient LearningCode2
MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency RegularizationCode1
SimMatch: Semi-supervised Learning with Similarity MatchingCode0
Class-Aware Contrastive Semi-Supervised LearningCode1
Global-Local Regularization Via Distributional RobustnessCode0
Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without SupervisionCode0
Debiased Self-Training for Semi-Supervised LearningCode1
Deep Reference Priors: What is the best way to pretrain a model?Code0
Contrastive Regularization for Semi-Supervised Learning0
Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet?Code0
Show:102550
← PrevPage 1 of 4Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SimCLR (ResNet-50 4×)Top 5 Accuracy92.6Unverified
2Rotation + VAT + Ent. Min.Top 5 Accuracy91.23Unverified
3SimCLR (ResNet-50 2×)Top 5 Accuracy91.2Unverified
4Mean Teacher (ResNeXt-152)Top 5 Accuracy90.89Unverified
5OBoW (ResNet-50)Top 5 Accuracy90.7Unverified
6R2-D2 (ResNet-18)Top 5 Accuracy90.48Unverified
7FixMatchTop 5 Accuracy89.13Unverified
8UDATop 5 Accuracy88.52Unverified
9SimCLR (ResNet-50)Top 5 Accuracy87.8Unverified
10DHO (ViT-Large)Top 1 Accuracy85.9Unverified
#ModelMetricClaimedVerifiedStatus
1DHO (ViT-Large)Top 1 Accuracy84.6Unverified
2OBoW (ResNet-50)Top 5 Accuracy82.9Unverified
3DHO (ViT-Base)Top 1 Accuracy81.6Unverified
4REACT (ViT-Large)Top 1 Accuracy81.6Unverified
5Semi-SST (ViT-Huge)Top 1 Accuracy80.7Unverified
6Meta Co-TrainingTop 1 Accuracy80.7Unverified
7Super-SST (ViT-Huge)Top 1 Accuracy80.3Unverified
8Semi-ViT (ViT-Huge)Top 1 Accuracy80Unverified
9Semi-ViT (ViT-Large)Top 1 Accuracy77.3Unverified
10Super-SST (ViT-Small distilled)Top 1 Accuracy76.9Unverified
#ModelMetricClaimedVerifiedStatus
1Γ-modelPercentage error20.4Unverified
2GANPercentage error15.59Unverified
3Bad GANPercentage error14.41Unverified
4Triple-GAN-V2 (CNN-13, no aug)Percentage error12.41Unverified
5Pi ModelPercentage error12.16Unverified
6SESEMI SSL (ConvNet)Percentage error11.65Unverified
7VATPercentage error11.36Unverified
8GLOT-DRPercentage error10.6Unverified
9VAT+EntMinPercentage error10.55Unverified
10Triple-GAN-V2 (CNN-13)Percentage error10.01Unverified
#ModelMetricClaimedVerifiedStatus
1Ⅱ-ModelPercentage error39.19Unverified
2SESEMI SSL (ConvNet)Percentage error38.7Unverified
3Temporal ensemblingPercentage error38.65Unverified
4R2-D2 (CNN-13)Percentage error32.87Unverified
5Dual Student (480)Percentage error32.77Unverified
6UPS (CNN-13)Percentage error32Unverified
7SHOT-VAEPercentage error25.3Unverified
8LiDAMPercentage error23.22Unverified
9EnAET (WRN-28-2-Large)Percentage error22.92Unverified
10FixMatch (RA, WRN-28-8)Percentage error22.6Unverified
#ModelMetricClaimedVerifiedStatus
1Ⅱ-ModelPercentage error53.12Unverified
2MixUpPercentage error47.43Unverified
3MeanTeacherPercentage error47.32Unverified
4VATPercentage error36.03Unverified
5LiDAMPercentage error19.17Unverified
6MixMatchPercentage error11.08Unverified
7RealMixPercentage error9.79Unverified
8EnAETPercentage error7.6Unverified
9ReMixMatchPercentage error6.27Unverified
10FixMatch+CRPercentage error5.04Unverified