Revisiting Network Perturbation for Semi-Supervised Semantic Segmentation

TL;DR

This paper introduces MLPMatch, a novel semi-supervised semantic segmentation framework that enhances network perturbation techniques and combines multiple perturbation levels, achieving state-of-the-art results on Pascal VOC and Cityscapes datasets.

Contribution

It proposes a new network perturbation approach and a volatile learning process, expanding weak-to-strong consistency regularization for unlabeled data in semi-supervised segmentation.

Findings

Achieves state-of-the-art performance on Pascal VOC.

Effective integration of network perturbation with consistency regularization.

Validated on Cityscapes dataset with strong results.

Abstract

In semi-supervised semantic segmentation (SSS), weak-to-strong consistency regularization techniques are widely utilized in recent works, typically combined with input-level and feature-level perturbations. However, the integration between weak-to-strong consistency regularization and network perturbation has been relatively rare. We note several problems with existing network perturbations in SSS that may contribute to this phenomenon. By revisiting network perturbations, we introduce a new approach for network perturbation to expand the existing weak-to-strong consistency regularization for unlabeled data. Additionally, we present a volatile learning process for labeled data, which is uncommon in existing research. Building upon previous work that includes input-level and feature-level perturbations, we present MLPMatch (Multi-Level-Perturbation Match), an easy-to-implement and efficient framework for semi-supervised semantic segmentation. MLPMatch has been validated on the Pascal VOC and Cityscapes datasets, achieving state-of-the-art performance. Code is available from https://github.com/LlistenL/MLPMatch.