DeepAI AI Chat
Log In Sign Up

PseudoSeg: Designing Pseudo Labels for Semantic Segmentation

by   Yuliang Zou, et al.

Recent advances in semi-supervised learning (SSL) demonstrate that a combination of consistency regularization and pseudo-labeling can effectively improve image classification accuracy in the low-data regime. Compared to classification, semantic segmentation tasks require much more intensive labeling costs. Thus, these tasks greatly benefit from data-efficient training methods. However, structured outputs in segmentation render particular difficulties (e.g., designing pseudo-labeling and augmentation) to apply existing SSL strategies. To address this problem, we present a simple and novel re-design of pseudo-labeling to generate well-calibrated structured pseudo labels for training with unlabeled or weakly-labeled data. Our proposed pseudo-labeling strategy is network structure agnostic to apply in a one-stage consistency training framework. We demonstrate the effectiveness of the proposed pseudo-labeling strategy in both low-data and high-data regimes. Extensive experiments have validated that pseudo labels generated from wisely fusing diverse sources and strong data augmentation are crucial to consistency training for segmentation. The source code is available at


page 5

page 16

page 17


Re-distributing Biased Pseudo Labels for Semi-supervised Semantic Segmentation: A Baseline Investigation

While self-training has advanced semi-supervised semantic segmentation, ...

Logic-induced Diagnostic Reasoning for Semi-supervised Semantic Segmentation

Recent advances in semi-supervised semantic segmentation have been heavi...

Few-shot Class-Incremental Semantic Segmentation via Pseudo-Labeling and Knowledge Distillation

We address the problem of learning new classes for semantic segmentation...

Inferior Alveolar Nerve Segmentation in CBCT images using Connectivity-Based Selective Re-training

Inferior Alveolar Nerve (IAN) canal detection in CBCT is an important st...

SPICE: Semantic Pseudo-labeling for Image Clustering

This paper presents SPICE, a Semantic Pseudo-labeling framework for Imag...

POPCORN: Progressive Pseudo-labeling with Consistency Regularization and Neighboring

Semi-supervised learning (SSL) uses unlabeled data to compensate for the...

Explainable COVID-19 Infections Identification and Delineation Using Calibrated Pseudo Labels

The upheaval brought by the arrival of the COVID-19 pandemic has continu...