Instance-specific 6-DoF Object Pose Estimation from Minimal Annotations

TL;DR

This paper introduces a rapid, instance-specific 6-DoF object pose estimation method that uses minimal manual annotations and leverages known camera poses and geometry to generate training data, suitable for familiar scenarios.

Contribution

The authors propose a novel pipeline that automates dataset generation and trains neural networks for accurate pose estimation with minimal manual labeling in known environments.

Findings

Accurately estimates object pose with minimal manual annotation.

Effective in scenarios with minor environmental variations.

Rapid training and deployment for specific object instances.

Abstract

In many robotic applications, the environment setting in which the 6-DoF pose estimation of a known, rigid object and its subsequent grasping is to be performed, remains nearly unchanging and might even be known to the robot in advance. In this paper, we refer to this problem as instance-specific pose estimation: the robot is expected to estimate the pose with a high degree of accuracy in only a limited set of familiar scenarios. Minor changes in the scene, including variations in lighting conditions and background appearance, are acceptable but drastic alterations are not anticipated. To this end, we present a method to rapidly train and deploy a pipeline for estimating the continuous 6-DoF pose of an object from a single RGB image. The key idea is to leverage known camera poses and rigid body geometry to partially automate the generation of a large labeled dataset. The dataset, along with sufficient domain randomization, is then used to supervise the training of deep neural networks for predicting semantic keypoints. Experimentally, we demonstrate the convenience and effectiveness of our proposed method to accurately estimate object pose requiring only a very small amount of manual annotation for training.