Binding Touch to Everything: Learning Unified Multimodal Tactile Representations
UniTouch aligns tactile signals with vision, language and audio embeddings, enabling zero-shot touch understanding.UniTouch 将触觉信号与视觉、语言、音频嵌入对齐，实现零样本触觉理解。
BibTeX
@misc{yang2024binding,
title = {{Binding Touch to Everything: Learning Unified Multimodal Tactile Representations}},
author = {Fengyu Yang and Chao Feng and Ziyang Chen and Hyoungseob Park and Daniel Wang and Yiming Dou and Ziyao Zeng and Xien Chen and Rit Gangopadhyay and Andrew Owens and Alex Wong},
year = {2024},
eprint = {2401.18084},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2401.18084},
}