GAN-based image-to-friction generation for tactile simulation of fabric material

Cai, Shaoyu; Zhao, Lu; Ban, Yuki; Narumi, Takuji; Liu, Yue; Zhu, Kening

doi:10.1016/j.cag.2021.09.007

Cited by 21 publications

(7 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These devices often use vibration, piezoelectricity, electrostatics, and ultrasound as driving methods to provide users with the most natural tactile interaction experience. Shaoyu Cai et al [5]. proposed an electric vibration-based tactile display.…”

Section: Embedded Tactile Feedback Devicementioning

confidence: 99%

Review: the development of tactile feedback devices

Zhu,

Yang,

Song

et al. 2024

Ninth International Symposium on Sensors, Mechatronics, and Automation System (ISSMAS 2023)

View full text Add to dashboard Cite

With the advent of the holographic communication era, the amount and types of information continue to increase. People urgently need to explore new methods of information interaction. Tactile interaction technology can produce tactile feedback, greatly enhancing multimedia interactivity and user immersion, and it is an indispensable key technology in future multisensory interactive communication. This paper will introduce the latest research progress of tactile feedback devices in the past 5 years. According to different installation methods, it will be divided into four types and will summarize the characteristics of the most representative tactile feedback devices in the past 5 years in a table. Subsequently, the paper will summarize the existing problems of each type of tactile feedback device and analyze the future development trends of this field. The application of haptic feedback technology is prospected in the end.

show abstract

Section: Embedded Tactile Feedback Devicementioning

confidence: 99%

Review: the development of tactile feedback devices

Zhu,

Yang,

Song

et al. 2024

Ninth International Symposium on Sensors, Mechatronics, and Automation System (ISSMAS 2023)

View full text Add to dashboard Cite

show abstract

“…The main focus of this study revolves around the utilization of transfer learning in generative models. Generative Adversarial Networks (GANs) have garnered significant attention in recent years, leading researchers from various domains to explore their potential in diverse tasks such as generating signals from image inputs [20]. This section provides an extensive overview of experiments conducted using two generative models: Restricted Boltzmann Machines (RBMs) and GANs, with a specific emphasis on their application as classifiers.…”

Section: Related Workmentioning

confidence: 99%

How Transfer Learning Is Used in Generative Models forImage Classification: Improved Accuracy

Ebrahimzadeh,

Sharif,

Banad

2024

Preprint

View full text Add to dashboard Cite

Recent breakthroughs in generative neural networks have paved the way for transformative capabilities,particularly in their capacity to generate novel data, notably in the realm of images. Theintegration of these models with the increasingly popular technique of transfer learning, designed forproficient feature extraction, holds the promise of enhancing overall performance. This paper delvesinto the exploration of employing generative models in conjunction with transfer learning methods forfeature extraction, with a specific focus on image classification tasks. Our investigation aims to scrutinizethe effectiveness of leveraging generative models alongside pre-trained models as feature extractorsin the context of image classification. To the best of our knowledge, our investigation is the first tolink transfer learning and generative models for a discriminative task under one roof. The proposedapproach undergoes rigorous evaluation on two distinct datasets, employing specific metrics to gaugethe model’s performance. The results exhibit a notable nearly 10% enhancement achieved through theintegration of generative models, underscoring their potential for achieving heightened accuracy inimage classification. These findings highlight significant advancements in image classification accuracy,surpassing the performance of conventional Artificial Neural Network (ANN) models.

show abstract

“…To eliminate complex steps of tactile data collection, Cai et al [22] use a generative model to synthesise the frictional signals from visual images, then the synthesised frictional signals are rendered on the haptic display instead of the real signal. However, the touch feeling of the object's surface relates to the roughness and slipperiness, and the above work [22] only considers frictional coefficients in a straight line but ignores the height disparities over the object's surface. In our proposed method, we use visual information and generative models to generate height map and frictional coefficients data, and combine them together for haptic rendering, for the first time.…”

Section: Tactile Signals Rendering On Electrovibration Haptic Displaymentioning

confidence: 99%