Transformer in Touch: A Survey

TL;DR

This survey reviews the emerging use of Transformer models in tactile perception, highlighting their core mechanisms, applications, and future research directions in tactile technology.

Contribution

It provides a comprehensive overview of Transformer applications in tactile tasks, summarizing methodologies, benchmarks, and design insights for the tactile community.

Findings

Transformers are effective in tactile object recognition.

They enable cross-modal generation in tactile systems.

Performance benchmarks show competitive results.

Abstract

The Transformer model, initially achieving significant success in the field of natural language processing, has recently shown great potential in the application of tactile perception. This review aims to comprehensively outline the application and development of Transformers in tactile technology. We first introduce the two fundamental concepts behind the success of the Transformer: the self-attention mechanism and large-scale pre-training. Then, we delve into the application of Transformers in various tactile tasks, including but not limited to object recognition, cross-modal generation, and object manipulation, offering a concise summary of the core methodologies, performance benchmarks, and design highlights. Finally, we suggest potential areas for further research and future work, aiming to generate more interest within the community, tackle existing challenges, and encourage the use of Transformer models in the tactile field.