# Landmarks Augmentation with Manifold-Barycentric Oversampling

arXiv: https://arxiv.org/abs/2104.00925 · HTML: https://mlsanity.com/paper/2104.00925/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 72 references that are in the map.
- First posted on arXiv 2 Apr 2021.

## Builds on (most cited first, 30 of 34)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Stacked Hourglass Networks for Human Pose Estimation](https://mlsanity.com/paper/1603.06937/) (2016)
- [Multimodal Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/1804.04732/) (2018)
- [Encoding in Style: a StyleGAN Encoder for Image-to-Image Translation](https://mlsanity.com/paper/2008.00951/) (2020)
- [Neural Spline Flows](https://mlsanity.com/paper/1906.04032/) (2019)
- [Generative Adversarial Networks: An Overview](https://mlsanity.com/paper/1710.07035/) (2017)
- [Albumentations: fast and flexible image augmentations](https://mlsanity.com/paper/1809.06839/) (2018)
- [The Effectiveness of Data Augmentation in Image Classification using Deep Learning](https://mlsanity.com/paper/1712.04621/) (2017)
- [Learning Generative Models with Sinkhorn Divergences](https://mlsanity.com/paper/1706.00292/) (2017)
- [DualGAN: Unsupervised Dual Learning for Image-to-Image Translation](https://mlsanity.com/paper/1704.02510/) (2017)
- [Fast Computation of Wasserstein Barycenters](https://mlsanity.com/paper/1310.4375/) (2013)
- [Dataset Augmentation in Feature Space](https://mlsanity.com/paper/1702.05538/) (2017)
- [Unsupervised Learning of Object Landmarks through Conditional Image Generation](https://mlsanity.com/paper/1806.07823/) (2018)
- [Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks](https://mlsanity.com/paper/1711.06753/) (2017)
- [Unsupervised Discovery of Object Landmarks as Structural Representations](https://mlsanity.com/paper/1804.04412/) (2018)
- [Unsupervised Part-Based Disentangling of Object Shape and Appearance](https://mlsanity.com/paper/1903.06946/) (2019)
- [Unsupervised learning of object landmarks by factorized spatial embeddings](https://mlsanity.com/paper/1705.02193/) (2017)
- [Self-supervised learning of a facial attribute embedding from video](https://mlsanity.com/paper/1808.06882/) (2018)
- [Improving Deep Learning using Generic Data Augmentation](https://mlsanity.com/paper/1708.06020/) (2017)
- [Debiased Sinkhorn barycenters](https://mlsanity.com/paper/2006.02575/) (2020)
- [Aggregation via Separation: Boosting Facial Landmark Detector with Semi-Supervised Style Translation](https://mlsanity.com/paper/1908.06440/) (2019)
- [Anomaly Detection in Medical Imaging with Deep Perceptual Autoencoders](https://mlsanity.com/paper/2006.13265/) (2020)
- [Automated Design of Deep Learning Methods for Biomedical Image Segmentation](https://mlsanity.com/paper/1904.08128/) (2019)
- [Scalable Nearest Neighbor Search for Optimal Transport](https://mlsanity.com/paper/1910.04126/) (2019)
- [A Survey on Face Data Augmentation](https://mlsanity.com/paper/1904.11685/) (2019)
- [Wasserstein barycenters can be computed in polynomial time in fixed dimension](https://mlsanity.com/paper/2006.08012/) (2020)
- [Unsupervised learning of object frames by dense equivariant image labelling](https://mlsanity.com/paper/1706.02932/) (2017)
- [WaveCycleGAN: Synthetic-to-natural speech waveform conversion using cycle-consistent adversarial networks](https://mlsanity.com/paper/1809.10288/) (2018)

## Built on by (most cited first, 4 of 4)

- [Align Your Intents: Offline Imitation Learning via Optimal Transport](https://mlsanity.com/paper/2402.13037/) (2024)
- [ENOT: Expectile Regularization for Fast and Accurate Training of Neural Optimal Transport](https://mlsanity.com/paper/2403.03777/) (2024)
- [Wasserstein Geodesic Generator for Conditional Distributions](https://mlsanity.com/paper/2308.10145/) (2023)
- [Simplicial SMOTE: Oversampling Solution to the Imbalanced Learning Problem](https://mlsanity.com/paper/2503.03418/) (2025)

## Abstract (arXiv)

The training of Generative Adversarial Networks (GANs) requires a large amount of data, stimulating the development of new augmentation methods to alleviate the challenge. Oftentimes, these methods either fail to produce enough new data or expand the dataset beyond the original manifold. In this paper, we propose a new augmentation method that guarantees to keep the new data within the original data manifold thanks to the optimal transport theory. The proposed algorithm finds cliques in the nearest-neighbors graph and, at each sampling iteration, randomly draws one clique to compute the Wasserstein barycenter with random uniform weights. These barycenters then become the new natural-looking elements that one could add to the dataset. We apply this approach to the problem of landmarks detection and augment the available annotation in both unpaired and in semi-supervised scenarios. Additionally, the idea is validated on cardiac data for the task of medical segmentation. Our approach reduces the overfitting and improves the quality metrics beyond the original data outcome and beyond the result obtained with popular modern augmentation methods.