GANSlider: How Users Control Generative Models for Images using Multiple Sliders with and without Feedforward Information

TL;DR

This study examines how multiple sliders and visual feedback influence user control over StyleGAN2, revealing tradeoffs between control complexity, visualization benefits, and task performance in image reconstruction.

Contribution

It provides empirical insights into UI design factors affecting user interaction with generative models, highlighting the impact of control dimensions and visualizations on usability.

Findings

More sliders increase task difficulty and user actions.

Visual feedforward improves goal-directed interaction.

No significant improvement in task speed or accuracy.

Abstract

We investigate how multiple sliders with and without feedforward visualizations influence users' control of generative models. In an online study (N=138), we collected a dataset of people interacting with a generative adversarial network (StyleGAN2) in an image reconstruction task. We found that more control dimensions (sliders) significantly increase task difficulty and user actions. Visual feedforward partly mitigates this by enabling more goal-directed interaction. However, we found no evidence of faster or more accurate task performance. This indicates a tradeoff between feedforward detail and implied cognitive costs, such as attention. Moreover, we found that visualizations alone are not always sufficient for users to understand individual control dimensions. Our study quantifies fundamental UI design factors and resulting interaction behavior in this context, revealing opportunities for improvement in the UI design for interactive applications of generative models. We close by discussing design directions and further aspects.