Bootstrapping Semantic Segmentation with Regional Contrast

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 semi-supervised 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. Code is available at https://github.com/lorenmt/reco.

Published at ICLR 2022. Project Page: https://shikun.io/projects/regional-contrast. Code: https://github.com/lorenmt/reco

References in corpus (2)

• DMT: Dynamic Mutual Training for Semi-Supervised Learning
• Unsupervised Learning of Dense Visual Representations

Cited by in corpus (4)

• LM-CPPF: Paraphrasing-Guided Data Augmentation for Contrastive Prompt-Based Few-Shot Fine-Tuning
• Leveraging Labelled Data Knowledge: A Cooperative Rectification Learning Network for Semi-supervised 3D Medical Image Segmentation
• Boundary-Refined Prototype Generation: A General End-to-End Paradigm for Semi-Supervised Semantic Segmentation
• Multi-dataset Pretraining: A Unified Model for Semantic Segmentation