OrthographicNet: A Deep Transfer Learning Approach for 3D Object Recognition in Open-Ended Domains

TL;DR

OrthographicNet is a CNN-based model that provides rotation- and scale-invariant 3D object recognition, enabling open-ended learning and real-time recognition for service robots in dynamic environments.

Contribution

It introduces OrthographicNet, a novel deep transfer learning approach that improves recognition accuracy, scalability, and few-shot learning capabilities in open-ended 3D object recognition.

Findings

Significant improvements over state-of-the-art methods.

Effective learning of new categories from few examples.

Real-time recognition demonstrated in real-world scenarios.

Abstract

Nowadays, service robots are appearing more and more in our daily life. For this type of robot, open-ended object category learning and recognition is necessary since no matter how extensive the training data used for batch learning, the robot might be faced with a new object when operating in a real-world environment. In this work, we present OrthographicNet, a Convolutional Neural Network (CNN)-based model, for 3D object recognition in open-ended domains. In particular, OrthographicNet generates a global rotation- and scale-invariant representation for a given 3D object, enabling robots to recognize the same or similar objects seen from different perspectives. Experimental results show that our approach yields significant improvements over the previous state-of-the-art approaches concerning object recognition performance and scalability in open-ended scenarios. Moreover, OrthographicNet demonstrates the capability of learning new categories from very few examples on-site. Regarding real-time performance, three real-world demonstrations validate the promising performance of the proposed architecture.