OBBSeg: Irregular Lesion Segmentation under Oriented Bounding Box Annotations

2026-07-07

AI Summary

This work addresses the challenge of scaling irregular lesion segmentation in medical images, hindered by the high cost of pixel-level annotations, by proposing OBBSeg—a weakly supervised segmentation method leveraging oriented bounding boxes (OBBs) as compact geometric supervision. To mitigate rectangular bias inherent in OBB-based supervision, the authors introduce a differentiable Mask-to-OBB loss and enhance semantic guidance through two novel modules: Prompt-driven Foreground Enhancement (PAFE) and Dynamic Background Filtering (DBFE). Comprehensive experiments across five imaging modalities and thirteen datasets demonstrate that OBBSeg substantially outperforms existing weakly supervised approaches and achieves performance approaching that of fully supervised models, confirming its efficiency and strong generalization capability.

Abstract

Pixel-level annotation remains a major bottleneck in medical image segmentation, making weak supervision an attractive yet under-constrained alternative. We propose OBBSeg, an intermediate supervision paradigm guided by Oriented Bounding Boxes (OBBs) that bridges the gap between full and weak supervision. By jointly encoding spatial extent and orientation, OBBs provide compact geometric supervision that better aligns with elongated or anisotropic lesions, reducing the ambiguity of coarse box annotations. To mitigate the inherent rectangular bias of OBBs, we introduce a Mask-to-OBB loss, a differentiable formulation that enforces geometric consistency between predicted masks and OBB regions. Furthermore, we incorporate prompt-driven semantic guidance through two complementary modules-PAFE and DBFE-which enhance foreground representation and suppress background interference. Extensive experiments on 13 datasets across 5 imaging modalities show that OBBSeg not only outperforms existing weakly supervised methods but also achieves performance comparable to fully supervised approaches, demonstrating its potential for efficient and scalable medical image segmentation. The code is available at https://github.com/StarLxc3/OBBSeg.

Problem

• medical image segmentation
• weak supervision
• irregular lesion
• oriented bounding box
• annotation bottleneck

Innovation

• Oriented Bounding Box
• Weakly Supervised Segmentation
• Mask-to-OBB Loss
• Prompt-driven Guidance
• Medical Image Segmentation