Hybrid Bayesian Eigenobjects: Combining Linear Subspace and Deep Network Methods for 3D Robot Vision

TL;DR

This paper presents HBEOs, a new 3D object representation that combines linear subspace and deep learning to enable robots to efficiently estimate pose, class, and shape of objects from a single view.

Contribution

The introduction of HBEOs, a hybrid model that improves 3D object understanding by integrating linear subspace methods with deep convolutional networks, eliminating the need for voxelization.

Findings

Significantly better accuracy in pose, class, and shape estimation.

Orders of magnitude faster inference times.

Effective on novel objects from single viewpoints.

Abstract

We introduce Hybrid Bayesian Eigenobjects (HBEOs), a novel representation for 3D objects designed to allow a robot to jointly estimate the pose, class, and full 3D geometry of a novel object observed from a single viewpoint in a single practical framework. By combining both linear subspace methods and deep convolutional prediction, HBEOs efficiently learn nonlinear object representations without directly regressing into high-dimensional space. HBEOs also remove the onerous and generally impractical necessity of input data voxelization prior to inference. We experimentally evaluate the suitability of HBEOs to the challenging task of joint pose, class, and shape inference on novel objects and show that, compared to preceding work, HBEOs offer dramatically improved performance in all three tasks along with several orders of magnitude faster runtime performance.