Tell me what this is: Few-Shot Incremental Object Learning by a Robot

TL;DR

This paper introduces a practical system enabling robots to incrementally learn new object categories from few examples, perform object organization tasks, and predict object arrangements, closely matching batch-trained performance.

Contribution

It applies a state-of-the-art few-shot incremental learning method to robotic object recognition and demonstrates its effectiveness in real-world tasks.

Findings

Achieves near-batch training performance in incremental learning

Enables robots to organize objects and predict arrangements

Demonstrates practical applicability in robotic tasks

Abstract

For many applications, robots will need to be incrementally trained to recognize the specific objects needed for an application. This paper presents a practical system for incrementally training a robot to recognize different object categories using only a small set of visual examples provided by a human. The paper uses a recently developed state-of-the-art method for few-shot incremental learning of objects. After learning the object classes incrementally, the robot performs a table cleaning task organizing objects into categories specified by the human. We also demonstrate the system's ability to learn arrangements of objects and predict missing or incorrectly placed objects. Experimental evaluations demonstrate that our approach achieves nearly the same performance as a system trained with all examples at one time (batch training), which constitutes a theoretical upper bound.