When Regression Meets Manifold Learning for Object Recognition and Pose Estimation

TL;DR

This paper introduces a multi-task learning framework combining manifold descriptor learning and pose regression for object recognition and pose estimation from depth images, achieving significant accuracy improvements.

Contribution

It presents a novel combined approach that integrates manifold learning with pose regression, enhancing discriminative view descriptors and pose estimation accuracy.

Findings

30% increase in relative pose accuracy

Improved discriminative view descriptors

Enhanced object recognition and pose retrieval

Abstract

In this work, we propose a method for object recognition and pose estimation from depth images using convolutional neural networks. Previous methods addressing this problem rely on manifold learning to learn low dimensional viewpoint descriptors and employ them in a nearest neighbor search on an estimated descriptor space. In comparison we create an efficient multi-task learning framework combining manifold descriptor learning and pose regression. By combining the strengths of manifold learning using triplet loss and pose regression, we could either estimate the pose directly reducing the complexity compared to NN search, or use learned descriptor for the NN descriptor matching. By in depth experimental evaluation of the novel loss function we observed that the view descriptors learned by the network are much more discriminative resulting in almost 30% increase regarding relative pose accuracy compared to related works. On the other hand, regarding directly regressed poses we obtained important improvement compared to simple pose regression. By leveraging the advantages of both manifold learning and regression tasks, we are able to improve the current state-of-the-art for object recognition and pose retrieval that we demonstrate through in depth experimental evaluation.