Polar Transformation Based Multiple Instance Learning Assisting Weakly Supervised Image Segmentation With Loose Bounding Box Annotations

TL;DR

This paper introduces a novel polar transformation-based multiple instance learning method to improve weakly supervised image segmentation with loose bounding boxes, demonstrating superior performance on medical datasets.

Contribution

It proposes a new polar transformation strategy combined with weighted smooth maximum approximation for weakly supervised segmentation with loose bounding boxes.

Findings

Achieved higher Dice coefficients on a medical dataset.

Demonstrated the effectiveness of polar transformation in weak supervision.

Provided open-source code for reproducibility.

Abstract

This study investigates weakly supervised image segmentation using loose bounding box supervision. It presents a multiple instance learning strategy based on polar transformation to assist image segmentation when loose bounding boxes are employed as supervision. In this strategy, weighted smooth maximum approximation is introduced to incorporate the observation that pixels closer to the origin of the polar transformation are more likely to belong to the object in the bounding box. The proposed approach was evaluated on a public medical dataset using Dice coefficient. The results demonstrate its superior performance. The codes are available at \url{https://github.com/wangjuan313/wsis-polartransform}.