Unsupervised Semantic Segmentation

Edit

3 papers with code • 5 benchmarks • 3 datasets

Models that learn to segment each image (i.e. cluster the pixels into their ground truth classes) without seeing the ground truth labels.

Benchmarks

Add a Result

Dataset	Best Model	Results
PASCAL VOC 2012 val	MaskContrast (Saliency) Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals	3
COCO-Stuff-15	IIC Invariant Information Clustering for Unsupervised Image Classification and Segmentation	1
COCO-Stuff-3	IIC Invariant Information Clustering for Unsupervised Image Classification and Segmentation	1
Potsdam	IIC Invariant Information Clustering for Unsupervised Image Classification and Segmentation	1
Potsdam-3	IIC Invariant Information Clustering for Unsupervised Image Classification and Segmentation	1

Paper	Code	Results	Date	Stars