TEXasGAN: Tactile Texture Exploration and Synthesis System Using Generative Adversarial Network

To create more realistic experiences in human-virtual object interactions, texture rendering has become a research hotspot in recent years. Different frequency components of designed vibrations can activate texture-related sensations due to similar receptors. However, designing specific vibrations for numerous real-world materials is impractical. Therefore, this study proposes a human-in-the-loop vibration generation model based on user preferences. To enable users to easily control the generation of vibration samples with large parameter spaces, we introduce an optimization model based on Differential Subspace Search (DSS) and Generative Adversarial Network (GAN). With DSS, users can use a one-dimensional slider to easily modify the high-dimensional latent space so that the GAN can generate desired vibrations. We trained the generative model using a open dataset of tactile vibration data and selected five types of vibrations as target samples for the generation experiment. Extensive user experiments were conducted using the generated and real samples. The results indicate that our system can generate distinguishable samples that match the target characteristics. Moreover, the results also reveal a correlation between subjects' ability to distinguish real samples and their ability to distinguish generated samples.