CheckerPose: Progressive Dense Keypoint Localization for Object Pose Estimation with Graph Neural Network

TL;DR

CheckerPose introduces a dense, progressive keypoint sampling method combined with graph neural networks to improve 6-DoF object pose estimation accuracy from RGB images, achieving state-of-the-art results.

Contribution

It proposes a novel dense keypoint sampling and correspondence refinement approach using binary codes and graph neural networks for enhanced pose estimation.

Findings

Achieves state-of-the-art accuracy on Linemod, Linemod-O, and YCB-V benchmarks.

Boosts the reliability and precision of dense correspondence methods.

Demonstrates the effectiveness of GNNs in modeling keypoint interactions for pose estimation.

Abstract

Estimating the 6-DoF pose of a rigid object from a single RGB image is a crucial yet challenging task. Recent studies have shown the great potential of dense correspondence-based solutions, yet improvements are still needed to reach practical deployment. In this paper, we propose a novel pose estimation algorithm named CheckerPose, which improves on three main aspects. Firstly, CheckerPose densely samples 3D keypoints from the surface of the 3D object and finds their 2D correspondences progressively in the 2D image. Compared to previous solutions that conduct dense sampling in the image space, our strategy enables the correspondence searching in a 2D grid (i.e., pixel coordinate). Secondly, for our 3D-to-2D correspondence, we design a compact binary code representation for 2D image locations. This representation not only allows for progressive correspondence refinement but also converts the correspondence regression to a more efficient classification problem. Thirdly, we adopt a graph neural network to explicitly model the interactions among the sampled 3D keypoints, further boosting the reliability and accuracy of the correspondences. Together, these novel components make CheckerPose a strong pose estimation algorithm. When evaluated on the popular Linemod, Linemod-O, and YCB-V object pose estimation benchmarks, CheckerPose clearly boosts the accuracy of correspondence-based methods and achieves state-of-the-art performances. Code is available at https://github.com/RuyiLian/CheckerPose.