Automatic Content-aware Non-Photorealistic Rendering of Images

TL;DR

This paper introduces a novel content-aware framework for non-photorealistic image rendering that separately manipulates salient and non-salient regions to achieve various visual effects like detail exaggeration and abstraction.

Contribution

The paper presents a new content-aware approach that enables automatic, seamless non-photorealistic rendering of images by independently processing salient and non-salient regions.

Findings

Effective separation of salient and non-salient regions

Seamless blending of processed image regions

Versatile application to multiple visual effects

Abstract

Non-photorealistic rendering techniques work on image features and often manipulate a set of characteristics such as edges and texture to achieve a desired depiction of the scene. Most computational photography methods decompose an image using edge preserving filters and work on the resulting base and detail layers independently to achieve desired visual effects. We propose a new approach for content-aware non-photorealistic rendering of images where we manipulate the visually salient and the non-salient regions separately. We propose a novel content-aware framework in order to render an image for applications such as detail exaggeration, artificial blurring and image abstraction. The processed regions of the image are blended seamlessly for all these applications. We demonstrate that content awareness of the proposed method leads to automatic generation of non-photorealistic rendering of the same image for the different applications mentioned above.