# CyclePose -- Leveraging Cycle-Consistency for Annotation-Free Nuclei Segmentation in Fluorescence Microscopy

arXiv: https://arxiv.org/abs/2503.11266 · HTML: https://mlsanity.com/paper/2503.11266/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 22 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Builds on (most cited first, 9 of 9)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Panoptic Segmentation](https://mlsanity.com/paper/1801.00868/) (2018)
- [Cell Detection with Star-convex Polygons](https://mlsanity.com/paper/1806.03535/) (2018)
- [DenoiSeg: Joint Denoising and Segmentation](https://mlsanity.com/paper/2005.02987/) (2020)
- [Three Dimensional Fluorescence Microscopy Image Synthesis and Segmentation](https://mlsanity.com/paper/1801.07198/) (2018)
- [Domain Adaptation-based Augmentation for Weakly Supervised Nuclei Detection](https://mlsanity.com/paper/1907.04681/) (2019)
- [AD-GAN: End-to-end Unsupervised Nuclei Segmentation with Aligned Disentangling Training](https://mlsanity.com/paper/2107.11022/) (2021)
- [Focus on Content not Noise: Improving Image Generation for Nuclei Segmentation by Suppressing Steganography in CycleGAN](https://mlsanity.com/paper/2308.01769/) (2023)

## Abstract (arXiv)

In recent years, numerous neural network architectures specifically designed for the instance segmentation of nuclei in microscopic images have been released. These models embed nuclei-specific priors to outperform generic architectures like U-Nets; however, they require large annotated datasets, which are often not available. Generative models (GANs, diffusion models) have been used to compensate for this by synthesizing training data. These two-stage approaches are computationally expensive, as first a generative model and then a segmentation model has to be trained. We propose CyclePose, a hybrid framework integrating synthetic data generation and segmentation training. CyclePose builds on a CycleGAN architecture, which allows unpaired translation between microscopy images and segmentation masks. We embed a segmentation model into CycleGAN and leverage a cycle consistency loss for self-supervision. Without annotated data, CyclePose outperforms other weakly or unsupervised methods on two public datasets. Code is available at https://github.com/jonasutz/CyclePose