PARASOL: Parametric Style Control for Diffusion Image Synthesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tarrés, Gemma Canet, Ruta, Dan, Bui, Tu, Collomosse, John
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911861784444928
author Tarrés, Gemma Canet
Ruta, Dan
Bui, Tu
Collomosse, John
author_facet Tarrés, Gemma Canet
Ruta, Dan
Bui, Tu
Collomosse, John
contents We propose PARASOL, a multi-modal synthesis model that enables disentangled, parametric control of the visual style of the image by jointly conditioning synthesis on both content and a fine-grained visual style embedding. We train a latent diffusion model (LDM) using specific losses for each modality and adapt the classifier-free guidance for encouraging disentangled control over independent content and style modalities at inference time. We leverage auxiliary semantic and style-based search to create training triplets for supervision of the LDM, ensuring complementarity of content and style cues. PARASOL shows promise for enabling nuanced control over visual style in diffusion models for image creation and stylization, as well as generative search where text-based search results may be adapted to more closely match user intent by interpolating both content and style descriptors.
format Preprint
id arxiv_https___arxiv_org_abs_2303_06464
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle PARASOL: Parametric Style Control for Diffusion Image Synthesis
Tarrés, Gemma Canet
Ruta, Dan
Bui, Tu
Collomosse, John
Computer Vision and Pattern Recognition
We propose PARASOL, a multi-modal synthesis model that enables disentangled, parametric control of the visual style of the image by jointly conditioning synthesis on both content and a fine-grained visual style embedding. We train a latent diffusion model (LDM) using specific losses for each modality and adapt the classifier-free guidance for encouraging disentangled control over independent content and style modalities at inference time. We leverage auxiliary semantic and style-based search to create training triplets for supervision of the LDM, ensuring complementarity of content and style cues. PARASOL shows promise for enabling nuanced control over visual style in diffusion models for image creation and stylization, as well as generative search where text-based search results may be adapted to more closely match user intent by interpolating both content and style descriptors.
title PARASOL: Parametric Style Control for Diffusion Image Synthesis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2303.06464