DoUnseen: Tuning-Free Class-Adaptive Object Detection of Unseen Objects for Robotic Grasping

TL;DR

This paper introduces DoUnseen, a tuning-free, class-adaptive object detection method for robotic grasping that can add new objects on the fly without retraining, addressing open-set segmentation challenges.

Contribution

It proposes a novel two-step segmentation approach combining unseen object segmentation networks with class-adaptive classifiers, enabling real-time addition of new object classes without fine-tuning.

Findings

Performance varies depending on environment and objects

Outperforms traditional trained models in open-set scenarios

Provides a practical solution for robotic grasping applications

Abstract

How can we segment varying numbers of objects where each specific object represents its own separate class? To make the problem even more realistic, how can we add and delete classes on the fly without retraining or fine-tuning? This is the case of robotic applications where no datasets of the objects exist or application that includes thousands of objects (E.g., in logistics) where it is impossible to train a single model to learn all of the objects. Most current research on object segmentation for robotic grasping focuses on class-level object segmentation (E.g., box, cup, bottle), closed sets (specific objects of a dataset; for example, YCB dataset), or deep learning-based template matching. In this work, we are interested in open sets where the number of classes is unknown, varying, and without pre-knowledge about the objects' types. We consider each specific object as its own separate class. Our goal is to develop an object detector that requires no fine-tuning and can add any object as a class just by capturing a few images of the object. Our main idea is to break the segmentation pipelines into two steps by combining unseen object segmentation networks cascaded by class-adaptive classifiers. We evaluate our class-adaptive object detector on unseen datasets and compare it to a trained Mask R-CNN on those datasets. The results show that the performance varies from practical to unsuitable depending on the environment setup and the objects being handled. The code is available in our DoUnseen library repository.