Diffusion idea exploration for art generation

TL;DR

This paper explores the use of diffusion models for creative art generation guided by text and sketches, showing promising initial results in producing novel images.

Contribution

It introduces the application of state-of-the-art diffusion models for cross-modal art generation, surpassing previous GAN, autoregressive, and VAE methods.

Findings

Diffusion models effectively generate creative images guided by text and sketches.

Initial experiments show promising qualitative results.

Diffusion models outperform prior generative techniques in this context.

Abstract

Cross-Modal learning tasks have picked up pace in recent times. With plethora of applications in diverse areas, generation of novel content using multiple modalities of data has remained a challenging problem. To address the same, various generative modelling techniques have been proposed for specific tasks. Novel and creative image generation is one important aspect for industrial application which could help as an arm for novel content generation. Techniques proposed previously used Generative Adversarial Network(GAN), autoregressive models and Variational Autoencoders (VAE) for accomplishing similar tasks. These approaches are limited in their capability to produce images guided by either text instructions or rough sketch images decreasing the overall performance of image generator. We used state of the art diffusion models to generate creative art by primarily leveraging text with additional support of rough sketches. Diffusion starts with a pattern of random dots and slowly converts that pattern into a design image using the guiding information fed into the model. Diffusion models have recently outperformed other generative models in image generation tasks using cross modal data as guiding information. The initial experiments for this task of novel image generation demonstrated promising qualitative results.