Identity-Aware U-Net: Fine-grained Cell Segmentation via Identity-Aware Representation Learning

TL;DR

The paper introduces Identity-Aware U-Net, a novel segmentation framework that combines spatial localization with discriminative identity learning to improve fine-grained cell segmentation, especially in challenging scenarios.

Contribution

It proposes a unified model with an auxiliary embedding branch and triplet-based metric learning to enhance discrimination among similar objects in dense segmentation tasks.

Findings

Improved segmentation accuracy on cell datasets, especially with similar contours.

Effective discrimination between morphologically similar objects.

Enhanced robustness in ambiguous boundary cases.

Abstract

Precise segmentation of objects with highly similar shapes remains a challenging problem in dense prediction, especially in scenarios with ambiguous boundaries, overlapping instances, and weak inter-instance visual differences. While conventional segmentation models are effective at localizing object regions, they often lack the discriminative capacity required to reliably distinguish a target object from morphologically similar distractors. In this work, we study fine-grained object segmentation from an identity-aware perspective and propose Identity-Aware U-Net (IAU-Net), a unified framework that jointly models spatial localization and instance discrimination. Built upon a U-Net-style encoder-decoder architecture, our method augments the segmentation backbone with an auxiliary embedding branch that learns discriminative identity representations from high-level features, while the main branch predicts pixel-accurate masks. To enhance robustness in distinguishing objects with near-identical contours or textures, we further incorporate triplet-based metric learning, which pulls target-consistent embeddings together and separates them from hard negatives with similar morphology. This design enables the model to move beyond category-level segmentation and acquire a stronger capability for precise discrimination among visually similar objects. Experiments on benchmarks including cell segmentation demonstrate promising results, particularly in challenging cases involving similar contours, dense layouts, and ambiguous boundaries.