Computer Vision
 
Edit

Semi-Supervised Image Classification

124 papers with code • 58 benchmarks • 13 datasets

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 )

Libraries

Use these libraries to find Semi-Supervised Image Classification models and implementations
facebookresearch/vissl
7 papers
3,230
lightly-ai/lightly
7 papers
2,756
Westlake-AI/openmixup
7 papers
574
vturrisi/solo-learn
6 papers
1,357
See all 16 libraries.

Datasets

Subtasks

Most implemented papers

Most implemented Social Latest No code

VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning

facebookresearch/vicreg • • NeurIPS 2021

Recent self-supervised methods for image representation learning are based on maximizing the agreement between embedding vectors from different views of the same image.

6
Paper
Code

Unsupervised Feature Learning via Non-Parametric Instance Discrimination

zhirongw/lemniscate.pytorch • • CVPR 2018

Neural net classifiers trained on data with annotated class labels can also capture apparent visual similarity among categories without being directed to do so.

4
Paper
Code

Interpolation Consistency Training for Semi-Supervised Learning

vikasverma1077/ICT • • 9 Mar 2019

We introduce Interpolation Consistency Training (ICT), a simple and computation efficient algorithm for training Deep Neural Networks in the semi-supervised learning paradigm.

4
Paper
Code

Data-Efficient Image Recognition with Contrastive Predictive Coding

philip-bachman/amdim-public • • ICML 2020

Human observers can learn to recognize new categories of images from a handful of examples, yet doing so with artificial ones remains an open challenge.

4
Paper
Code

Large Scale Adversarial Representation Learning

lukemelas/unsupervised-image-segmentation • • NeurIPS 2019

We extensively evaluate the representation learning and generation capabilities of these BigBiGAN models, demonstrating that these generation-based models achieve the state of the art in unsupervised representation learning on ImageNet, as well as in unconditional image generation.

4
Paper
Code

Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples

facebookresearch/suncet • • ICCV 2021

This paper proposes a novel method of learning by predicting view assignments with support samples (PAWS).

4
Paper
Code

With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations

lightly-ai/lightly • • ICCV 2021

On semi-supervised learning benchmarks we improve performance significantly when only 1% ImageNet labels are available, from 53. 8% to 56. 5%.

4
Paper
Code

FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning

torchssl/torchssl • • 15 May 2022

Semi-supervised Learning (SSL) has witnessed great success owing to the impressive performances brought by various methods based on pseudo labeling and consistency regularization.

4
Paper
Code

USB: A Unified Semi-supervised Learning Benchmark for Classification

microsoft/semi-supervised-learning • • 12 Aug 2022

We further provide the pre-trained versions of the state-of-the-art neural models for CV tasks to make the cost affordable for further tuning.

4
Paper
Code

ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring

google-research/mixmatch • • 21 Nov 2019

Distribution alignment encourages the marginal distribution of predictions on unlabeled data to be close to the marginal distribution of ground-truth labels.

3
Paper
Code
Contact us on: hello@paperswithcode.com . Papers With Code is a free resource with all data licensed under CC-BY-SA.
Terms Data policy Cookies policy