Abstract:Designing attractive images for object display can be a time-consuming and skill-intensive process. The emergence of advanced algorithms, particularly the Diffusion Model, has made it possible to synthesize attractive images using AI. However, the existing diffusion models are mostly used to generate entire images and lack control over specific objects for object display. Here, to the best of our knowledge, we pioneers to extend the application of the diffusion model to synthesize novel images for specific objects. By encoding the input images of objects into NeRF representation and synthesizing the desired backgrounds using diffusion models with the input of rendered object images and text prompts, our method can generate 3D aware object display images at arbitrary angles and arbitrary backgrounds. We have conducted extensive experiments to demonstrate that our method is capable of generating high-quality and photo-realistic images, which are > 6 times faster than the conventional photomontage approach. Moreover, our generated images have higher compositional scores, image quality scores, and aesthetics scores in our user experiments. By significantly reducing the need for human effort and producing higher quality generated images, our approach opens up exciting possibilities for creating versatile novel images of specific objects.

Realization of Cartoon Effects Based on Non-Photorealistic Rendering

PencilArt: A Chromatic Penciling Style Generation Framework.

Real-Time Ocean Water Animation in Cartoon Style

Caster: Cartoon Style Transfer Via Dynamic Cartoon Style Casting

Selective Image Abstraction

Generating Abstract Paintings in Kandinsky Style

Real-time Cartoon Water Animation

Computer-Aided Rendering of Chinese Meticulous Figure Paintings

The Algorithms And Principles Of Non-Photorealistic Graphics Artistic Rendering And Cartoon Animation Introduction

Style Conversion of Cartoon Animation

AI supported computer-generated pen-and-ink illustration

Automatic Image-Based Pencil Sketch Rendering

Perspectives Of Non-Photorealistic Computer Graphics

An adaptive cartoon-like stylization for color video in real time

Real-Time Painterly Abstraction by Pseudo Morphological Filtering

Novel 3D-Aware Composition Images Synthesis for Object Display with Diffusion Model.

Computer generation of cartoon ripples and shimmering

Based on Human-like Approach

An Art-Oriented Pixelation Method for Cartoon Images

PTGCF: Printing Texture Guided Color Fusion for Impressionism Oil Painting Style Rendering

Cartoon-like Stylization of Video for Real-Time Applications.