# Automatic Semantic Style Transfer using Deep Convolutional Neural Networks and Soft Masks

arXiv: https://arxiv.org/abs/1708.09641 · HTML: https://mlsanity.com/paper/1708.09641/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Aug 2017.

## Built on by (most cited first, 7 of 7)

- [CoARF: Controllable 3D Artistic Style Transfer for Radiance Fields](https://mlsanity.com/paper/2404.14967/) (2024)
- [Say No to Freeloader: Protecting Intellectual Property of Your Deep Model](https://mlsanity.com/paper/2408.13161/) (2024)
- [SCSA: A Plug-and-Play Semantic Continuous-Sparse Attention for Arbitrary Semantic Style Transfer](https://mlsanity.com/paper/2503.04119/) (2025)
- [MixSyn: Learning Composition and Style for Multi-Source Image Synthesis](https://mlsanity.com/paper/2111.12705/) (2021)
- [Neural Policy Style Transfer](https://mlsanity.com/paper/2402.00677/) (2024)
- [Improving Masked Style Transfer using Blended Partial Convolution](https://mlsanity.com/paper/2508.05769/) (2025)
- [Arbitrary Handwriting Image Style Transfer](https://mlsanity.com/paper/2201.05346/) (2022)

## Abstract (arXiv)

This paper presents an automatic image synthesis method to transfer the style of an example image to a content image. When standard neural style transfer approaches are used, the textures and colours in different semantic regions of the style image are often applied inappropriately to the content image, ignoring its semantic layout, and ruining the transfer result. In order to reduce or avoid such effects, we propose a novel method based on automatically segmenting the objects and extracting their soft semantic masks from the style and content images, in order to preserve the structure of the content image while having the style transferred. Each soft mask of the style image represents a specific part of the style image, corresponding to the soft mask of the content image with the same semantics. Both the soft masks and source images are provided as multichannel input to an augmented deep CNN framework for style transfer which incorporates a generative Markov random field (MRF) model. Results on various images show that our method outperforms the most recent techniques.