Touch In The Wild: Visuo-Tactile Demonstrations for Fine-Grained Manipulation

A wide range of visuo-tactile demonstrations collected using a portable visuo-tactile gripper supports the paper 'Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper'. The dataset spans both structured lab settings and unstructured real-world environments and is divided into three major categories. It was authored by binghaohuang-robot and last updated on 2026-01-21.

Use Cases

• Training robotic manipulation models based on visuo-tactile demonstrations.
• Benchmarking fine-grained control algorithms based on data from structured and unstructured environments.
• Developing portable robotic gripper systems based on the described visuo-tactile sensor data.

Strengths

• Data collection spans both structured lab and unstructured real-world environments, suggesting ecological diversity.
• The dataset is divided into three major categories, indicating a structured organization.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

binghaohuang-robot

Collection Method

Collected using a portable visuo-tactile gripper.

Freshness

Last updated 2026-01-21 19:18:43; freshness should be verified.