UNIC Dataset Part 1

Introduction

This repository provides the dataset used in UNIC: Learning Unified Multimodal Extrinsic Contact Estimation.
The dataset contains multimodal sensory data collected for training and evaluating the UNIC framework. It is intended to support research on extrinsic contact estimation and contact-rich robotic manipulation.
Details of the model architecture, training procedure, and evaluation protocols are provided in the released UNIC code repository. Please refer to the repository for all training and implementation details.

At a Glance
•    The size of the unzipped dataset is ~98GB
•    The dataset is divided into two (Part 1 and Part 2)

Other Resources
The corresponding code can be found at https://github.com/merlresearch/unic.

Citation

If you use the UNIC dataset in your research, please cite our contribution:

@inproceedings{xu2026unic,
 	author = {Xu, Zhengtong and Shirai, Yuki},
  	title = {UNIC: Learning Unified Multimodal Extrinsic Contact Estimation},
  	booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
  	year = {2026}
}

License

The UNIC dataset is released under CC-BY-SA-4.0 license.

All data:

Created by Mitsubishi Electric Research Laboratories (MERL), 2026
 
SPDX-License-Identifier: CC-BY-SA-4.0