# Disentangled Unsupervised Image Translation via Restricted Information Flow

arXiv: https://arxiv.org/abs/2111.13279 · HTML: https://mlsanity.com/paper/2111.13279/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 27 references that are in the map.
- First posted on arXiv 26 Nov 2021.

## Builds on (most cited first, 26 of 26)

- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [StarGAN v2: Diverse Image Synthesis for Multiple Domains](https://mlsanity.com/paper/1912.01865/) (2019)
- [Least Squares Generative Adversarial Networks](https://mlsanity.com/paper/1611.04076/) (2016)
- [Multimodal Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/1804.04732/) (2018)
- [MaskGAN: Towards Diverse and Interactive Facial Image Manipulation](https://mlsanity.com/paper/1907.11922/) (2019)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [Unsupervised Image-to-Image Translation Networks](https://mlsanity.com/paper/1703.00848/) (2017)
- [A Neural Algorithm of Artistic Style](https://mlsanity.com/paper/1508.06576/) (2015)
- [DRIT++: Diverse Image-to-Image Translation via Disentangled Representations](https://mlsanity.com/paper/1905.01270/) (2019)
- [Texture Networks: Feed-forward Synthesis of Textures and Stylized Images](https://mlsanity.com/paper/1603.03417/) (2016)
- [Fine-Grained Head Pose Estimation Without Keypoints](https://mlsanity.com/paper/1710.00925/) (2017)
- [PersonLab: Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Model](https://mlsanity.com/paper/1803.08225/) (2018)
- [Deep Learning for Deepfakes Creation and Detection: A Survey](https://mlsanity.com/paper/1909.11573/) (2019)
- [Generative Models for Effective ML on Private, Decentralized Datasets](https://mlsanity.com/paper/1911.06679/) (2019)
- [Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data](https://mlsanity.com/paper/1802.10151/) (2018)
- [Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting](https://mlsanity.com/paper/1906.09531/) (2019)
- [CycleGAN, a Master of Steganography](https://mlsanity.com/paper/1712.02950/) (2017)
- [Unsupervised Learning for Real-World Super-Resolution](https://mlsanity.com/paper/1909.09629/) (2019)
- [PixColor: Pixel Recursive Colorization](https://mlsanity.com/paper/1705.07208/) (2017)
- [Unsupervised Diverse Colorization via Generative Adversarial Networks](https://mlsanity.com/paper/1702.06674/) (2017)
- [Unsupervised Sparse Dirichlet-Net for Hyperspectral Image Super-Resolution](https://mlsanity.com/paper/1804.05042/) (2018)
- [Adversarial Self-Defense for Cycle-Consistent GANs](https://mlsanity.com/paper/1908.01517/) (2019)
- [Domain Intersection and Domain Difference](https://mlsanity.com/paper/1908.11628/) (2019)
- [TwoStreamVAN: Improving Motion Modeling in Video Generation](https://mlsanity.com/paper/1812.01037/) (2018)
- [Evaluation of Correctness in Unsupervised Many-to-Many Image Translation](https://mlsanity.com/paper/2103.15727/) (2021)

## Built on by (most cited first, 1 of 1)

- [COLA: A Benchmark for Compositional Text-to-image Retrieval](https://mlsanity.com/paper/2305.03689/) (2023)

## Abstract (arXiv)

Unsupervised image-to-image translation methods aim to map images from one domain into plausible examples from another domain while preserving structures shared across two domains. In the many-to-many setting, an additional guidance example from the target domain is used to determine domain-specific attributes of the generated image. In the absence of attribute annotations, methods have to infer which factors are specific to each domain from data during training. Many state-of-art methods hard-code the desired shared-vs-specific split into their architecture, severely restricting the scope of the problem. In this paper, we propose a new method that does not rely on such inductive architectural biases, and infers which attributes are domain-specific from data by constraining information flow through the network using translation honesty losses and a penalty on the capacity of domain-specific embedding. We show that the proposed method achieves consistently high manipulation accuracy across two synthetic and one natural dataset spanning a wide variety of domain-specific and shared attributes.