Skip to Main content Skip to Navigation
Conference papers

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

Abstract : 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
Document type :
Conference papers
Complete list of metadata

https://hal.telecom-paris.fr/hal-03108387
Contributor : Stéphane Lathuilière Connect in order to contact the contributor
Submitted on : Wednesday, January 13, 2021 - 10:35:04 AM
Last modification on : Tuesday, October 19, 2021 - 11:15:25 AM

Links full text

Identifiers

  • HAL Id : hal-03108387, version 1
  • ARXIV : 2101.03806

Collections

Citation

The-Phuc Nguyen, Stéphane Lathuilière, Elisa Ricci. Multi-Domain Image-to-Image Translation with Adaptive Inference Graph. International Conference on Pattern Recognition (ICPR 2020), Jan 2021, Milano, France. ⟨hal-03108387⟩

Share

Metrics

Record views

31