Exploring Spatial Diversity for Region-based Active Learning

TL;DR

This paper introduces a spatial diversity-aware region-based active learning framework for semantic segmentation, reducing annotation costs while maintaining high accuracy, and demonstrating superior performance on Cityscapes and PASCAL VOC datasets.

Contribution

It proposes incorporating local spatial diversity into active learning for semantic segmentation, improving selection strategies and reducing labeling effort.

Findings

Achieves 95% of fully supervised performance with only 5-9% labeled pixels.

Incorporating spatial diversity improves active learning effectiveness.

Outperforms existing region-based active learning methods.

Abstract

State-of-the-art methods for semantic segmentation are based on deep neural networks trained on large-scale labeled datasets. Acquiring such datasets would incur large annotation costs, especially for dense pixel-level prediction tasks like semantic segmentation. We consider region-based active learning as a strategy to reduce annotation costs while maintaining high performance. In this setting, batches of informative image regions instead of entire images are selected for labeling. Importantly, we propose that enforcing local spatial diversity is beneficial for active learning in this case, and to incorporate spatial diversity along with the traditional active selection criterion, e.g., data sample uncertainty, in a unified optimization framework for region-based active learning. We apply this framework to the Cityscapes and PASCAL VOC datasets and demonstrate that the inclusion of spatial diversity effectively improves the performance of uncertainty-based and feature diversity-based active learning methods. Our framework achieves $95\%$ performance of fully supervised methods with only $5-9\%$ of the labeled pixels, outperforming all state-of-the-art region-based active learning methods for semantic segmentation.