Multiple Interactions Made Easy (MIME): Large Scale Demonstrations Data for Imitation

TL;DR

This paper introduces MIME, the largest dataset of 8,260 human-robot demonstrations across 20 diverse manipulation tasks, enabling advances in visual imitation, trajectory prediction, and multi-task robotic learning.

Contribution

The paper presents MIME, a large-scale, multi-task robotic demonstration dataset with videos and trajectories, and explores methods for mapping video features to robot trajectories.

Findings

Dataset covers 20 diverse tasks including stacking and pouring

Two approaches evaluated for trajectory prediction from video

Dataset aims to foster research in imitation and multi-task learning

Abstract

In recent years, we have seen an emergence of data-driven approaches in robotics. However, most existing efforts and datasets are either in simulation or focus on a single task in isolation such as grasping, pushing or poking. In order to make progress and capture the space of manipulation, we would need to collect a large-scale dataset of diverse tasks such as pouring, opening bottles, stacking objects etc. But how does one collect such a dataset? In this paper, we present the largest available robotic-demonstration dataset (MIME) that contains 8260 human-robot demonstrations over 20 different robotic tasks (https://sites.google.com/view/mimedataset). These tasks range from the simple task of pushing objects to the difficult task of stacking household objects. Our dataset consists of videos of human demonstrations and kinesthetic trajectories of robot demonstrations. We also propose to use this dataset for the task of mapping 3rd person video features to robot trajectories. Furthermore, we present two different approaches using this dataset and evaluate the predicted robot trajectories against ground-truth trajectories. We hope our dataset inspires research in multiple areas including visual imitation, trajectory prediction, and multi-task robotic learning.