VTouch++: A Multimodal Dataset with Vision-Based Tactile Enhancement for Bimanual Manipulation

TL;DR

VTouch++ introduces a comprehensive multimodal dataset with vision-based tactile data, systematic task design, and scalable data collection to advance bimanual manipulation research.

Contribution

It provides a high-fidelity tactile dataset, a systematic task framework, and automated data collection pipelines for contact-rich manipulation tasks.

Findings

Effective cross-modal retrieval demonstrated

Real-robot evaluation shows practical applicability

Generalizable inference across robots and tasks

Abstract

Embodied intelligence has advanced rapidly in recent years; however, bimanual manipulation-especially in contact-rich tasks remains challenging. This is largely due to the lack of datasets with rich physical interaction signals, systematic task organization, and sufficient scale. To address these limitations, we introduce the VTOUCH dataset. It leverages vision based tactile sensing to provide high-fidelity physical interaction signals, adopts a matrix-style task design to enable systematic learning, and employs automated data collection pipelines covering real-world, demand-driven scenarios to ensure scalability. To further validate the effectiveness of the dataset, we conduct extensive quantitative experiments on cross-modal retrieval as well as real-robot evaluation. Finally, we demonstrate real-world performance through generalizable inference across multiple robots, policies, and tasks.