# LOTS of Fashion! Multi-Conditioning for Image Generation via Sketch-Text Pairing

Source: [LOTS of Fashion! Multi-Conditioning for Image Generation via Sketch-Text Pairing](<https://www.sota2.com/research/paper/68fa7b251e209131cea86d5d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:33:44.062Z

Authors: Federico Girella, Davide Talon, Ziyue Liu, Zanxi Ruan, Yiming Wang, Marco Cristani

Published: 2025-07-30T12:48:29.000Z

## Abstract

Fashion design is a complex creative process that blends visual and textual expressions. Designers convey ideas through sketches, which define spatial structure and design elements, and textual descriptions, capturing material, texture, and stylistic details. In this paper, we present LOcalized Text and Sketch for fashion image generation (LOTS), an approach for compositional sketch-text based generation of complete fashion outlooks. LOTS leverages a global description with paired localized sketch + text information for conditioning and introduces a novel step-based merging strategy for diffusion adaptation. First, a Modularized Pair-Centric representation encodes sketches and text into a shared latent space while preserving independent localized features; then, a Diffusion Pair Guidance phase integrates both local and global conditioning via attention-based guidance within the diffusion model's multi-step denoising process. To validate our method, we build on Fashionpedia to release Sketchy, the first fashion dataset where multiple text-sketch pairs are provided per image. Quantitative results show LOTS achieves state-of-the-art image generation performance on both global and localized metrics, while qualitative examples and a human evaluation study highlight its unprecedented level of design customization.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Sketch-to-Image Generation on Sketchy In the Wild](<https://www.sota2.com/research/sota/sketch-to-image-generation-on-sketchy-in-the-wild.md>): Sketch-to-Image Generation; Sketchy In the Wild; FID: 1.19.

- [Attribute Localization on Sketchy Human Sketches (Human Evaluation)](<https://www.sota2.com/research/sota/attribute-localization-on-sketchy-human-sketches-human-evaluation.md>): Attribute Localization; Sketchy Human Sketches (Human Evaluation); Precision: 81.3.

- [Structural Alignment on Sketchy Human Sketches (Human Evaluation)](<https://www.sota2.com/research/sota/structural-alignment-on-sketchy-human-sketches-human-evaluation.md>): Structural alignment; Sketchy Human Sketches (Human Evaluation); Preference Rate: 0.423.

- [Sketch-to-image on Sketchy (test)](<https://www.sota2.com/research/sota/sketch-to-image-on-sketchy-test.md>): Sketch-to-image; Sketchy (test); FID: 0.79.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2507.22627>)

[PDF](<https://arxiv.org/pdf/2507.22627.pdf>)