# Multi-Domain Image-to-Image Translation with Adaptive Inference Graph

arXiv: https://arxiv.org/abs/2101.03806 · HTML: https://mlsanity.com/paper/2101.03806/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 26 references that are in the map.
- First posted on arXiv 11 Jan 2021.

## Builds on (most cited first, 23 of 23)

- [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)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs](https://mlsanity.com/paper/1711.11585/) (2017)
- [StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation](https://mlsanity.com/paper/1711.09020/) (2017)
- [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)
- [First Order Motion Model for Image Animation](https://mlsanity.com/paper/2003.00196/) (2020)
- [DRIT++: Diverse Image-to-Image Translation via Disentangled Representations](https://mlsanity.com/paper/1905.01270/) (2019)
- [Animating Arbitrary Objects via Deep Motion Transfer](https://mlsanity.com/paper/1812.08861/) (2018)
- [Convolutional Networks with Adaptive Inference Graphs](https://mlsanity.com/paper/1711.11503/) (2017)
- [Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis](https://mlsanity.com/paper/1903.05628/) (2019)
- [Conditional Channel Gated Networks for Task-Aware Continual Learning](https://mlsanity.com/paper/2004.00070/) (2020)
- [Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping](https://mlsanity.com/paper/1809.05852/) (2018)
- [Refine and Distill: Exploiting Cycle-Inconsistency and Knowledge Distillation for Unsupervised Monocular Depth Estimation](https://mlsanity.com/paper/1903.04202/) (2019)
- [3D Guided Fine-Grained Face Manipulation](https://mlsanity.com/paper/1902.08900/) (2019)
- [ComboGAN: Unrestrained Scalability for Image Domain Translation](https://mlsanity.com/paper/1712.06909/) (2017)
- [Appearance and Pose-Conditioned Human Image Generation using Deformable GANs](https://mlsanity.com/paper/1905.00007/) (2019)
- [Monocular Depth Estimation using Multi-Scale Continuous CRFs as Sequential Deep Networks](https://mlsanity.com/paper/1803.00891/) (2018)
- [Art2Real: Unfolding the Reality of Artworks via Semantically-Aware Image-to-Image Translation](https://mlsanity.com/paper/1811.10666/) (2018)

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

- [Physics-informed Guided Disentanglement in Generative Networks](https://mlsanity.com/paper/2107.14229/) (2021)
- [FedStein: Enhancing Multi-Domain Federated Learning Through James-Stein Estimator](https://mlsanity.com/paper/2410.03499/) (2024)
- [Map Feature Perception Metric for Map Generation Quality Assessment and Loss Optimization](https://mlsanity.com/paper/2503.23370/) (2025)
- [Med-2D SegNet: A Light Weight Deep Neural Network for Medical 2D Image Segmentation](https://mlsanity.com/paper/2504.14715/) (2025)

## Abstract (arXiv)

In this work, we address the problem of multi-domain image-to-image translation with particular attention paid to computational cost. In particular, current state of the art models require a large and deep model in order to handle the visual diversity of multiple domains. In a context of limited computational resources, increasing the network size may not be possible. Therefore, we propose to increase the network capacity by using an adaptive graph structure. At inference time, the network estimates its own graph by selecting specific sub-networks. Sub-network selection is implemented using Gumbel-Softmax in order to allow end-to-end training. This approach leads to an adjustable increase in number of parameters while preserving an almost constant computational cost. Our evaluation on two publicly available datasets of facial and painting images shows that our adaptive strategy generates better images with fewer artifacts than literature methods