IPGAN: Identity-Preservation Generative Adversarial Network for unsupervised photo-to-caricature translation

作者：

Highlights：

• We propose a novel IPGAN for unsupervised photo-to-caricature translation.

• We propose an identity preservation loss to preserve identity of original photos.

• Experiments indicate that our IPGAN outperforms the existing state-of-the-arts.

• Our IPGAN can also be applied to caricature-to-photo translation.

摘要

•We propose a novel IPGAN for unsupervised photo-to-caricature translation.•We propose an identity preservation loss to preserve identity of original photos.•Experiments indicate that our IPGAN outperforms the existing state-of-the-arts.•Our IPGAN can also be applied to caricature-to-photo translation.

论文关键词：Photo-to-caricature translation,Generative adversarial networks,Image-to-image translation,Style transfer,Image warping

论文评审过程：Received 16 April 2021, Revised 7 January 2022, Accepted 14 January 2022, Available online 25 January 2022, Version of Record 5 February 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.108223