Skip to main content

ORIGINAL RESEARCH article

Front. Artif. Intell.
Sec. AI in Food, Agriculture and Water
Volume 7 - 2024 | doi: 10.3389/frai.2024.1498956

Dynamic-Budget Superpixel Active Learning for Semantic Segmentation

Provisionally accepted
  • Computer Science, University of Saskatchewan, Saskatoon, Saskatchewan, Canada

The final, formatted version of the article will be published soon.

    Active learning can significantly decrease the labelling cost of deep learning workflows by prioritizing the limited labelling budget to high-impact data points that have the highest positive impact on model accuracy. Active learning is especially useful for semantic segmentation tasks where we can selectively label only a few high-impact regions within these high-impact images. Most established regional active learning algorithms deploy a static-budget querying strategy where a fixed percentage of regions are queried in each image. A static budget could result in over-or under-labelling images as the number of high-impact regions in each image can vary. In this paper, we present a novel dynamic-budget superpixel querying strategy that can query the optimal numbers of high-uncertainty superpixels in an image to improve the querying efficiency of regional active learning algorithms designed for semantic segmentation. For two distinct datasets, we show that by allowing a dynamic budget for each image, the active learning algorithm is more effective compared to static-budget querying at the same low total labelling budget. We investigate both low-and high-budget scenarios and the impact of superpixel size on our dynamic active learning scheme. In a low-budget scenario, our dynamic-budget querying outperforms static-budget querying by 5.6% mIoU on a specialized agriculture field image dataset and 2.4% mIoU on Cityscapes.

    Keywords: Dynamic-budget Querying, Superpixel, Regional Querying, Active Learning, Semantic segmentation

    Received: 19 Sep 2024; Accepted: 16 Dec 2024.

    Copyright: © 2024 Wang and Stavness. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Ian Stavness, Computer Science, University of Saskatchewan, Saskatoon, S7N 5A2, Saskatchewan, Canada

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.