# vMFNet: Compositionality Meets Domain-generalised Segmentation

arXiv: https://arxiv.org/abs/2206.14538 · HTML: https://mlsanity.com/paper/2206.14538/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 43 references that are in the map.
- First posted on arXiv 29 Jun 2022.

## Builds on (most cited first, 29 of 29)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Domain Generalization by Solving Jigsaw Puzzles](https://mlsanity.com/paper/1903.06864/) (2019)
- [Domain Generalization via Model-Agnostic Learning of Semantic Features](https://mlsanity.com/paper/1910.13580/) (2019)
- [FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space](https://mlsanity.com/paper/2103.06030/) (2021)
- [Emergence of Invariance and Disentanglement in Deep Representations](https://mlsanity.com/paper/1706.01350/) (2017)
- [FSDR: Frequency Space Domain Randomization for Domain Generalization](https://mlsanity.com/paper/2103.02370/) (2021)
- [Deep learning for cardiac image segmentation: A review](https://mlsanity.com/paper/1911.03723/) (2019)
- [Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains](https://mlsanity.com/paper/2007.02035/) (2020)
- [Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization](https://mlsanity.com/paper/2009.12829/) (2020)
- [Test-Time Adaptable Neural Networks for Robust Medical Image Segmentation](https://mlsanity.com/paper/2004.04668/) (2020)
- [Learning to Compose Visual Relations](https://mlsanity.com/paper/2111.09297/) (2021)
- [Disentangled Representation Learning in Cardiac Image Analysis](https://mlsanity.com/paper/1903.09467/) (2019)
- [Enhancing Pseudo Label Quality for Semi-Supervised Domain-Generalized Medical Image Segmentation](https://mlsanity.com/paper/2201.08657/) (2022)
- [Learning Compositional Representations for Few-Shot Recognition](https://mlsanity.com/paper/1812.09213/) (2018)
- [Compositional Convolutional Neural Networks: A Deep Architecture with Innate Robustness to Partial Occlusion](https://mlsanity.com/paper/2003.04490/) (2020)
- [Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentation](https://mlsanity.com/paper/2106.13292/) (2021)
- [Compositional Transformers for Scene Generation](https://mlsanity.com/paper/2111.08960/) (2021)
- [Robust Instance Segmentation through Reasoning about Multi-Object Occlusion](https://mlsanity.com/paper/2012.02107/) (2020)
- [AtrialGeneral: Domain Generalization for Left Atrial Segmentation of Multi-Center LGE MRIs](https://mlsanity.com/paper/2106.08727/) (2021)
- [Domain Composition and Attention for Unseen-Domain Generalizable Medical Image Segmentation](https://mlsanity.com/paper/2109.08852/) (2021)
- [Disentangled Representations for Domain-generalized Cardiac Segmentation](https://mlsanity.com/paper/2008.11514/) (2020)
- [Measuring the Biases and Effectiveness of Content-Style Disentanglement](https://mlsanity.com/paper/2008.12378/) (2020)
- [Cooperative Training and Latent Space Data Augmentation for Robust Medical Image Segmentation](https://mlsanity.com/paper/2107.01079/) (2021)
- [Anatomy of Domain Shift Impact on U-Net Layers in MRI Segmentation](https://mlsanity.com/paper/2107.04914/) (2021)
- [Re-using Adversarial Mask Discriminators for Test-time Training under Distribution Shifts](https://mlsanity.com/paper/2108.11926/) (2021)
- [Stop Throwing Away Discriminators! Re-using Adversaries for Test-Time Training](https://mlsanity.com/paper/2108.12280/) (2021)

## Built on by (most cited first, 11 of 11)

- [Towards Generic Semi-Supervised Framework for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/2310.11320/) (2023)
- [RaffeSDG: Random Frequency Filtering enabled Single-source Domain Generalization for Medical Image Segmentation](https://mlsanity.com/paper/2405.01228/) (2024)
- [Compositionally Equivariant Representation Learning](https://mlsanity.com/paper/2306.07783/) (2023)
- [FedSemiDG: Domain Generalized Federated Semi-supervised Medical Image Segmentation](https://mlsanity.com/paper/2501.07378/) (2025)
- [S&D Messenger: Exchanging Semantic and Domain Knowledge for Generic Semi-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2407.07763/) (2024)
- [Dual-supervised Asymmetric Co-training for Semi-supervised Medical Domain Generalization](https://mlsanity.com/paper/2509.20785/) (2025)
- [Compositional Representation Learning for Brain Tumour Segmentation](https://mlsanity.com/paper/2310.06562/) (2023)
- [Enhancing Cross-Modal Medical Image Segmentation through Compositionality](https://mlsanity.com/paper/2408.11733/) (2024)
- [Casual Inference via Style Bias Deconfounding for Domain Generalization](https://mlsanity.com/paper/2503.16852/) (2025)
- [Diverse Teaching and Label Propagation for Generic Semi-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2508.08549/) (2025)
- [SNR-Adaptive Unified Diffusion for Multi-Task Medical Image Segmentation](https://mlsanity.com/paper/2607.03103/) (2026)

## Abstract (arXiv)

Training medical image segmentation models usually requires a large amount of labeled data. By contrast, humans can quickly learn to accurately recognise anatomy of interest from medical (e.g. MRI and CT) images with some limited guidance. Such recognition ability can easily generalise to new images from different clinical centres. This rapid and generalisable learning ability is mostly due to the compositional structure of image patterns in the human brain, which is less incorporated in medical image segmentation. In this paper, we model the compositional components (i.e. patterns) of human anatomy as learnable von-Mises-Fisher (vMF) kernels, which are robust to images collected from different domains (e.g. clinical centres). The image features can be decomposed to (or composed by) the components with the composing operations, i.e. the vMF likelihoods. The vMF likelihoods tell how likely each anatomical part is at each position of the image. Hence, the segmentation mask can be predicted based on the vMF likelihoods. Moreover, with a reconstruction module, unlabeled data can also be used to learn the vMF kernels and likelihoods by recombining them to reconstruct the input image. Extensive experiments show that the proposed vMFNet achieves improved generalisation performance on two benchmarks, especially when annotations are limited. Code is publicly available at: https://github.com/vios-s/vMFNet.