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

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Standard

Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space. / Grabe, Imke; Zhu, Jichen; Aguirrezabal Zabaleta, Manex.

International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar). Springer, Cham, 2022. p. 84-100.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Harvard

Grabe, I, Zhu, J & Aguirrezabal Zabaleta, M 2022, Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space. in International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar). Springer, Cham, pp. 84-100. <https://link.springer.com/chapter/10.1007/978-3-031-03789-4_6>

APA

Grabe, I., Zhu, J., & Aguirrezabal Zabaleta, M. (2022). Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space. In International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar) (pp. 84-100). https://link.springer.com/chapter/10.1007/978-3-031-03789-4_6

Vancouver

Grabe I, Zhu J, Aguirrezabal Zabaleta M. Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space. In International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar). Springer, Cham. 2022. p. 84-100

Author

Grabe, Imke ; Zhu, Jichen ; Aguirrezabal Zabaleta, Manex. / Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space. International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar). Springer, Cham, 2022. pp. 84-100

Bibtex

@inproceedings{6af9733674404685b2c57cdfaf9b7597,
title = "Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space",
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{\textquoteright}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.",
author = "Imke Grabe and Jichen Zhu and {Aguirrezabal Zabaleta}, Manex",
year = "2022",
language = "English",
pages = "84--100",
booktitle = "International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar)",

}

RIS

TY - GEN

T1 - Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space

AU - Grabe, Imke

AU - Zhu, Jichen

AU - Aguirrezabal Zabaleta, Manex

PY - 2022

Y1 - 2022

N2 - 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.

AB - 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.

M3 - Article in proceedings

SP - 84

EP - 100

BT - International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar)

CY - Springer, Cham

ER -

ID: 306304026