iCub World: Friendly Robots Help Building Good Vision Data-Sets

TL;DR

The paper introduces the iCub World dataset, a new object recognition dataset collected via human-robot interaction with the iCub robot, highlighting its utility for testing recognition systems and its accessibility for research.

Contribution

This work presents a novel dataset acquired through a human-robot interaction scheme using the iCub robot, facilitating rapid data collection and annotation for object recognition research.

Findings

Dataset challenges current recognition systems

Rapid data acquisition and annotation method

Publicly available dataset for research

Abstract

In this paper we present and start analyzing the iCub World data-set, an object recognition data-set, we acquired using a Human-Robot Interaction (HRI) scheme and the iCub humanoid robot platform. Our set up allows for rapid acquisition and annotation of data with corresponding ground truth. While more constrained in its scopes -- the iCub world is essentially a robotics research lab -- we demonstrate how the proposed data-set poses challenges to current recognition systems. The iCubWorld data-set is publicly available. The data-set can be downloaded from: http://www.iit.it/en/projects/data-sets.html.