Cell Instance Segmentation via Multi-Task Image-to-Image Schrödinger Bridge

2026-04-14

AI Summary

This work addresses the limitations of existing cell instance segmentation methods, which rely on deterministic predictions and post-processing and struggle to explicitly model the global structure of instance masks. The authors propose a novel approach that reformulates the task as a distribution-based multi-task image-to-image generation problem by introducing the Schrödinger bridge framework for the first time in this domain. The method incorporates boundary-aware inverse distance map supervision and enables stable predictions through deterministic inference. Notably, it achieves state-of-the-art or comparable performance on PanNuke without requiring SAM pretraining or post-processing, and demonstrates exceptional robustness under few-shot settings on MoNuSeg, significantly enhancing both segmentation quality and generalization capability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Existing cell instance segmentation pipelines typically combine deterministic predictions with post-processing, which imposes limited explicit constraints on the global structure of instance masks. In this work, we propose a multi-task image-to-image Schrödinger Bridge framework that formulates instance segmentation as a distribution-based image-to-image generation problem. Boundary-aware supervision is integrated through a reverse distance map, and deterministic inference is employed to produce stable predictions. Experimental results on the PanNuke dataset demonstrate that the proposed method achieves competitive or superior performance without relying on SAM pre-training or additional post-processing. Additional results on the MoNuSeg dataset show robustness under limited training data. These findings indicate that Schrödinger Bridge-based image-to-image generation provides an effective framework for cell instance segmentation.

Problem

• cell instance segmentation
• global structure
• instance masks
• image-to-image generation

Innovation

• Schrödinger Bridge
• cell instance segmentation
• image-to-image generation
• boundary-aware supervision
• multi-task learning