Bootstrapping Semantic Segmentation with Regional Contrast

Shikun Liu, Shuaifeng Zhi, Edward Johns, and Andrew Davison

Published at ICLR 2022

[Paper]          [Code]          [Blog]          [BibTex]

Abstract

We present ReCo, a contrastive learning framework designed at a regional level to assist learning in semantic segmentation. ReCo performs semi-supervised or supervised pixel-level contrastive learning on a sparse set of hard negative pixels, with minimal additional memory footprint. ReCo is easy to implement, being built on top of off-the-shelf segmentation networks, and consistently improves performance in both semisupervised and supervised semantic segmentation methods, achieving smoother segmentation boundaries and faster convergence. The strongest effect is in semi-supervised learning with very few labels. With ReCo, we achieve high quality semantic segmentation models, requiring only 5 examples of each semantic class. 

reco