Beyond Pixels: Semi-Supervised Semantic Segmentation with a Multi-scale Patch-based Multi-Label Classifier

TL;DR

This paper introduces a multi-scale patch-based multi-label classifier (MPMC) that enhances semi-supervised semantic segmentation by incorporating patch-level contextual information, improving performance across various datasets.

Contribution

The paper proposes MPMC, a novel plug-in module for semi-supervised segmentation frameworks that uses patch-level supervision and adaptive pseudo-label weighting.

Findings

MPMC improves segmentation accuracy across multiple datasets.

Integration of MPMC enhances existing semi-supervised methods.

MPMC effectively reduces noise from pseudo-label supervision.

Abstract

Incorporating pixel contextual information is critical for accurate segmentation. In this paper, we show that an effective way to incorporate contextual information is through a patch-based classifier. This patch classifier is trained to identify classes present within an image region, which facilitates the elimination of distractors and enhances the classification of small object segments. Specifically, we introduce Multi-scale Patch-based Multi-label Classifier (MPMC), a novel plug-in module designed for existing semi-supervised segmentation (SSS) frameworks. MPMC offers patch-level supervision, enabling the discrimination of pixel regions of different classes within a patch. Furthermore, MPMC learns an adaptive pseudo-label weight, using patch-level classification to alleviate the impact of the teacher's noisy pseudo-label supervision the student. This lightweight module can be integrated into any SSS framework, significantly enhancing their performance. We demonstrate the efficacy of our proposed MPMC by integrating it into four SSS methodologies and improving them across two natural image and one medical segmentation dataset, notably improving the segmentation results of the baselines across all the three datasets.