ControlGUI: Guiding Generative GUI Exploration through Perceptual Visual Flow

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Garg, Aryan, Jiang, Yue, Oulasvirta, Antti
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912985847431168
author Garg, Aryan
Jiang, Yue
Oulasvirta, Antti
author_facet Garg, Aryan
Jiang, Yue
Oulasvirta, Antti
contents During the early stages of interface design, designers need to produce multiple sketches to explore a design space. Design tools often fail to support this critical stage, because they insist on specifying more details than necessary. Although recent advances in generative AI have raised hopes of solving this issue, in practice they fail because expressing loose ideas in a prompt is impractical. In this paper, we propose a diffusion-based approach to the low-effort generation of interface sketches. It breaks new ground by allowing flexible control of the generation process via three types of inputs: A) prompts, B) wireframes, and C) visual flows. The designer can provide any combination of these as input at any level of detail, and will get a diverse gallery of low-fidelity solutions in response. The unique benefit is that large design spaces can be explored rapidly with very little effort in input-specification. We present qualitative results for various combinations of input specifications. Additionally, we demonstrate that our model aligns more accurately with these specifications than other models.
format Preprint
id arxiv_https___arxiv_org_abs_2502_03330
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ControlGUI: Guiding Generative GUI Exploration through Perceptual Visual Flow
Garg, Aryan
Jiang, Yue
Oulasvirta, Antti
Human-Computer Interaction
Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
During the early stages of interface design, designers need to produce multiple sketches to explore a design space. Design tools often fail to support this critical stage, because they insist on specifying more details than necessary. Although recent advances in generative AI have raised hopes of solving this issue, in practice they fail because expressing loose ideas in a prompt is impractical. In this paper, we propose a diffusion-based approach to the low-effort generation of interface sketches. It breaks new ground by allowing flexible control of the generation process via three types of inputs: A) prompts, B) wireframes, and C) visual flows. The designer can provide any combination of these as input at any level of detail, and will get a diverse gallery of low-fidelity solutions in response. The unique benefit is that large design spaces can be explored rapidly with very little effort in input-specification. We present qualitative results for various combinations of input specifications. Additionally, we demonstrate that our model aligns more accurately with these specifications than other models.
title ControlGUI: Guiding Generative GUI Exploration through Perceptual Visual Flow
topic Human-Computer Interaction
Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2502.03330