Deep Feature Rotation for Multimodal Image Style Transfer

TL;DR

This paper introduces Deep Feature Rotation (DFR), a simple and efficient method for multimodal image style transfer that generates diverse stylized outputs by augmenting intermediate feature embeddings through rotation.

Contribution

The paper presents DFR, a novel technique for style transfer that enables multiple stylized outputs from a single content and style image pair with minimal computational overhead.

Findings

DFR produces diverse stylized images effectively.

The method achieves comparable quality to complex style transfer approaches.

Visualization confirms the versatility of feature rotations.

Abstract

Recently, style transfer is a research area that attracts a lot of attention, which transfers the style of an image onto a content target. Extensive research on style transfer has aimed at speeding up processing or generating high-quality stylized images. Most approaches only produce an output from a content and style image pair, while a few others use complex architectures and can only produce a certain number of outputs. In this paper, we propose a simple method for representing style features in many ways called Deep Feature Rotation (DFR), while not only producing diverse outputs but also still achieving effective stylization compared to more complex methods. Our approach is representative of the many ways of augmentation for intermediate feature embedding without consuming too much computational expense. We also analyze our method by visualizing output in different rotation weights. Our code is available at https://github.com/sonnguyen129/deep-feature-rotation.