Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space

TL;DR

This paper introduces a method that uses evolutionary search and Gaussian mixture models within a GAN's latent space to generate fashion designs matching specific styles, enhancing style control in fashion image generation.

Contribution

It proposes a novel approach combining genetic algorithms and Gaussian mixture models to guide GANs in generating style-specific fashion designs.

Findings

Successfully generates fashion images matching target styles.

Demonstrates the effectiveness of evolutionary search in latent space.

Provides a new direction for style-coherent fashion design generation.

Abstract

This paper presents a novel approach for guiding a Generative Adversarial Network trained on the FashionGen dataset to generate designs corresponding to target fashion styles. Finding the latent vectors in the generator's latent space that correspond to a style is approached as an evolutionary search problem. A Gaussian mixture model is applied to identify fashion styles based on the higher-layer representations of outfits in a clothing-specific attribute prediction model. Over generations, a genetic algorithm optimizes a population of designs to increase their probability of belonging to one of the Gaussian mixture components or styles. Showing that the developed system can generate images of maximum fitness visually resembling certain styles, our approach provides a promising direction to guide the search for style-coherent designs.