# Keep it Simple: Image Statistics Matching for Domain Adaptation

arXiv: https://arxiv.org/abs/2005.12551 · HTML: https://mlsanity.com/paper/2005.12551/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 May 2020.

## Built on by (most cited first, 17 of 17)

- [Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation](https://mlsanity.com/paper/2101.10979/) (2021)
- [MeGA-CDA: Memory Guided Attention for Category-Aware Unsupervised Domain Adaptive Object Detection](https://mlsanity.com/paper/2103.04224/) (2021)
- [Unsupervised Domain Adaptation of Object Detectors: A Survey](https://mlsanity.com/paper/2105.13502/) (2021)
- [Fast Differentiable Matrix Square Root and Inverse Square Root](https://mlsanity.com/paper/2201.12543/) (2022)
- [Fast Differentiable Matrix Square Root](https://mlsanity.com/paper/2201.08663/) (2022)
- [GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates](https://mlsanity.com/paper/2304.07750/) (2023)
- [SynRS3D: A Synthetic Dataset for Global 3D Semantic Understanding from Monocular Remote Sensing Imagery](https://mlsanity.com/paper/2406.18151/) (2024)
- [Closer Look at the Uncertainty Estimation in Semantic Segmentation under Distributional Shift](https://mlsanity.com/paper/2106.00076/) (2021)
- [Using Unreliable Pseudo-Labels for Label-Efficient Semantic Segmentation](https://mlsanity.com/paper/2306.02314/) (2023)
- [AdaStereo: An Efficient Domain-Adaptive Stereo Matching Approach](https://mlsanity.com/paper/2112.04974/) (2021)
- [Foreground-Aware Stylization and Consensus Pseudo-Labeling for Domain Adaptation of First-Person Hand Segmentation](https://mlsanity.com/paper/2107.02718/) (2021)
- [Orthogonal SVD Covariance Conditioning and Latent Disentanglement](https://mlsanity.com/paper/2212.05599/) (2022)
- [Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality](https://mlsanity.com/paper/2207.02119/) (2022)
- [From Web Data to Real Fields: Low-Cost Unsupervised Domain Adaptation for Agricultural Robots](https://mlsanity.com/paper/2410.23906/) (2024)
- [Genetic Learning for Designing Sim-to-Real Data Augmentations](https://mlsanity.com/paper/2403.06786/) (2024)
- [SegDesicNet: Lightweight Semantic Segmentation in Remote Sensing with Geo-Coordinate Embeddings for Domain Adaptation](https://mlsanity.com/paper/2503.08290/) (2025)
- [Towards High-Resolution Alignment and Super-Resolution of Multi-Sensor Satellite Imagery](https://mlsanity.com/paper/2507.23150/) (2025)

## Abstract (arXiv)

Applying an object detector, which is neither trained nor fine-tuned on data close to the final application, often leads to a substantial performance drop. In order to overcome this problem, it is necessary to consider a shift between source and target domains. Tackling the shift is known as Domain Adaptation (DA). In this work, we focus on unsupervised DA: maintaining the detection accuracy across different data distributions, when only unlabeled images are available of the target domain. Recent state-of-the-art methods try to reduce the domain gap using an adversarial training strategy which increases the performance but at the same time the complexity of the training procedure. In contrast, we look at the problem from a new perspective and keep it simple by solely matching image statistics between source and target domain. We propose to align either color histograms or mean and covariance of the source images towards the target domain. Hence, DA is accomplished without architectural add-ons and additional hyper-parameters. The benefit of the approaches is demonstrated by evaluating different domain shift scenarios on public data sets. In comparison to recent methods, we achieve state-of-the-art performance using a much simpler procedure for the training. Additionally, we show that applying our techniques significantly reduces the amount of synthetic data needed to learn a general model and thus increases the value of simulation.