Bayesian Active Learning for Semantic Segmentation

TL;DR

This paper presents a Bayesian active learning framework for semantic segmentation that efficiently selects informative pixels for labeling, significantly reducing annotation effort while maintaining high accuracy.

Contribution

It introduces a novel pixel-level Bayesian uncertainty measure based on Balanced Entropy, scalable and analytically computable, improving active learning for segmentation tasks.

Findings

Achieves supervised-level mIoU with fewer labeled pixels

Outperforms previous active learning models on benchmark datasets

Scales linearly with closed-form computation

Abstract

Fully supervised training of semantic segmentation models is costly and challenging because each pixel within an image needs to be labeled. Therefore, the sparse pixel-level annotation methods have been introduced to train models with a subset of pixels within each image. We introduce a Bayesian active learning framework based on sparse pixel-level annotation that utilizes a pixel-level Bayesian uncertainty measure based on Balanced Entropy (BalEnt) [84]. BalEnt captures the information between the models' predicted marginalized probability distribution and the pixel labels. BalEnt has linear scalability with a closed analytical form and can be calculated independently per pixel without relational computations with other pixels. We train our proposed active learning framework for Cityscapes, Camvid, ADE20K and VOC2012 benchmark datasets and show that it reaches supervised levels of mIoU using only a fraction of labeled pixels while outperforming the previous state-of-the-art active learning models with a large margin.