Pixel Contrastive-Consistent Semi-Supervised Semantic Segmentation

TL;DR

This paper introduces PC2Seg, a semi-supervised semantic segmentation method that combines label-space consistency and feature-space contrastive learning, achieving state-of-the-art results on multiple datasets.

Contribution

The paper proposes a novel semi-supervised segmentation approach integrating pixel-level consistency and contrastive learning with negative sampling techniques.

Findings

Achieves state-of-the-art performance on VOC, Cityscapes, and COCO datasets.

Effectively addresses false negative noise in pixel contrastive loss.

Improves segmentation regularities through combined loss functions.

Abstract

We present a novel semi-supervised semantic segmentation method which jointly achieves two desiderata of segmentation model regularities: the label-space consistency property between image augmentations and the feature-space contrastive property among different pixels. We leverage the pixel-level L2 loss and the pixel contrastive loss for the two purposes respectively. To address the computational efficiency issue and the false negative noise issue involved in the pixel contrastive loss, we further introduce and investigate several negative sampling techniques. Extensive experiments demonstrate the state-of-the-art performance of our method (PC2Seg) with the DeepLab-v3+ architecture, in several challenging semi-supervised settings derived from the VOC, Cityscapes, and COCO datasets.