Unbiased Subclass Regularization for Semi-Supervised Semantic Segmentation

by   Dayan Guan, et al.
Nanyang Technological University

Semi-supervised semantic segmentation learns from small amounts of labelled images and large amounts of unlabelled images, which has witnessed impressive progress with the recent advance of deep neural networks. However, it often suffers from severe class-bias problem while exploring the unlabelled images, largely due to the clear pixel-wise class imbalance in the labelled images. This paper presents an unbiased subclass regularization network (USRN) that alleviates the class imbalance issue by learning class-unbiased segmentation from balanced subclass distributions. We build the balanced subclass distributions by clustering pixels of each original class into multiple subclasses of similar sizes, which provide class-balanced pseudo supervision to regularize the class-biased segmentation. In addition, we design an entropy-based gate mechanism to coordinate learning between the original classes and the clustered subclasses which facilitates subclass regularization effectively by suppressing unconfident subclass predictions. Extensive experiments over multiple public benchmarks show that USRN achieves superior performance as compared with the state-of-the-art.


page 1

page 3

page 6


Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation

We propose a novel deep neural network architecture for semi-supervised ...

Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization

Semi-supervised semantic segmentation requires the model to effectively ...

Frequentist-Bayes Hybrid Covariance Estimationfor Unfolding Problems

In this paper we present a frequentist-Bayesian hybrid method for estima...

Generalized Unbiased Scene Graph Generation

Existing Unbiased Scene Graph Generation (USGG) methods only focus on ad...

Semi-Supervised Semantic Segmentation of Vessel Images using Leaking Perturbations

Semantic segmentation based on deep learning methods can attain appealin...

SemiCurv: Semi-Supervised Curvilinear Structure Segmentation

Recent work on curvilinear structure segmentation has mostly focused on ...

Imbalance Trouble: Revisiting Neural-Collapse Geometry

Neural Collapse refers to the remarkable structural properties character...

Code Repositories

Please sign up or login with your details

Forgot password? Click here to reset