6D Object Pose Estimation using Keypoints and Part Affinity Fields

TL;DR

This paper introduces a two-step method for 6D object pose estimation from RGB images using keypoints and Part Affinity Fields, leveraging CNNs inspired by human pose estimation techniques.

Contribution

It adapts Part Affinity Fields for object pose estimation, improving keypoint assembly and accuracy over heatmap-only methods, with class-specific models outperforming multi-class models.

Findings

Achieves accuracy comparable to state-of-the-art methods on YCB-Video dataset.

Using PAFs improves keypoint assembly over heatmaps.

Class-specific models outperform multi-class models.

Abstract

The task of 6D object pose estimation from RGB images is an important requirement for autonomous service robots to be able to interact with the real world. In this work, we present a two-step pipeline for estimating the 6 DoF translation and orientation of known objects. Keypoints and Part Affinity Fields (PAFs) are predicted from the input image adopting the OpenPose CNN architecture from human pose estimation. Object poses are then calculated from 2D-3D correspondences between detected and model keypoints via the PnP-RANSAC algorithm. The proposed approach is evaluated on the YCB-Video dataset and achieves accuracy on par with recent methods from the literature. Using PAFs to assemble detected keypoints into object instances proves advantageous over only using heatmaps. Models trained to predict keypoints of a single object class perform significantly better than models trained for several classes.