Weakly Supervised Image Segmentation Beyond Tight Bounding Box Annotations

TL;DR

This paper introduces a novel polar transformation based multiple instance learning method for weakly supervised image segmentation that performs well with both tight and loose bounding box annotations, improving robustness and accuracy.

Contribution

It extends previous MIL approaches by integrating polar transformation, enabling effective segmentation with loose bounding boxes, which are easier to acquire than tight ones.

Findings

Achieves state-of-the-art performance across various bounding box precisions.

Demonstrates robustness to errors in loose bounding box annotations.

Effective on multiple public datasets with high dice coefficients.

Abstract

Weakly supervised image segmentation approaches in the literature usually achieve high segmentation performance using tight bounding box supervision and decrease the performance greatly when supervised by loose bounding boxes. However, compared with loose bounding box, it is much more difficult to acquire tight bounding box due to its strict requirements on the precise locations of the four sides of the box. To resolve this issue, this study investigates whether it is possible to maintain good segmentation performance when loose bounding boxes are used as supervision. For this purpose, this work extends our previous parallel transformation based multiple instance learning (MIL) for tight bounding box supervision by integrating an MIL strategy based on polar transformation to assist image segmentation. The proposed polar transformation based MIL formulation works for both tight and loose bounding boxes, in which a positive bag is defined as pixels in a polar line of a bounding box with one endpoint located inside the object enclosed by the box and the other endpoint located at one of the four sides of the box. Moreover, a 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 box. The proposed approach was evaluated on two public datasets using dice coefficient when bounding boxes at different precision levels were considered in the experiments. The results demonstrate that the proposed approach achieves state-of-the-art performance for bounding boxes at all precision levels and is robust to mild and moderate errors in the loose bounding box annotations. The codes are available at \url{https://github.com/wangjuan313/wsis-beyond-tightBB}.