Artistic Enhancement and Style Transfer of Image Edges using Directional Pseudo-coloring

TL;DR

This paper proposes a method called directional pseudo-coloring to artistically enhance and stylize image edges by coloring gradient magnitudes based on their directions, improving visual quality and enabling style transfer.

Contribution

It introduces a novel directional pseudo-coloring technique for image gradients and a style transfer pipeline that applies learned color maps to edges.

Findings

Enhanced visual quality of image edges through pseudo-coloring.

Effective style transfer of edge colors from style images.

Open-source code available for implementation.

Abstract

Computing the gradient of an image is a common step in computer vision pipelines. The image gradient quantifies the magnitude and direction of edges in an image and is used in creating features for downstream machine learning tasks. Typically, the image gradient is represented as a grayscale image. This paper introduces directional pseudo-coloring, an approach to color the image gradient in a deliberate and coherent manner. By pseudo-coloring the image gradient magnitude with the image gradient direction, we can enhance the visual quality of image edges and achieve an artistic transformation of the original image. Additionally, we present a simple style transfer pipeline which learns a color map from a style image and then applies that color map to color the edges of a content image through the directional pseudo-coloring technique. Code for the algorithms and experiments is available at https://github.com/shouvikmani/edge-colorizer.