ES6D: A Computation Efficient and Symmetry-Aware 6D Pose Regression Framework

TL;DR

This paper introduces ES6D, a fast and symmetry-aware 6D pose regression framework that directly estimates object poses from RGB-D images using a simple fully convolutional network, effectively handling symmetric objects and outperforming existing methods.

Contribution

The paper proposes a novel symmetry-invariant loss function and a simple, efficient architecture for 6D pose estimation from RGB-D data, improving accuracy and computational efficiency.

Findings

Superior accuracy on YCB-Video and T-LESS datasets.

Low computational cost compared to existing methods.

Effective handling of symmetric object ambiguities.

Abstract

In this paper, a computation efficient regression framework is presented for estimating the 6D pose of rigid objects from a single RGB-D image, which is applicable to handling symmetric objects. This framework is designed in a simple architecture that efficiently extracts point-wise features from RGB-D data using a fully convolutional network, called XYZNet, and directly regresses the 6D pose without any post refinement. In the case of symmetric object, one object has multiple ground-truth poses, and this one-to-many relationship may lead to estimation ambiguity. In order to solve this ambiguity problem, we design a symmetry-invariant pose distance metric, called average (maximum) grouped primitives distance or A(M)GPD. The proposed A(M)GPD loss can make the regression network converge to the correct state, i.e., all minima in the A(M)GPD loss surface are mapped to the correct poses. Extensive experiments on YCB-Video and T-LESS datasets demonstrate the proposed framework's substantially superior performance in top accuracy and low computational cost.