Large-scale 6D Object Pose Estimation Dataset for Industrial Bin-Picking

TL;DR

This paper introduces a comprehensive large-scale dataset for 6D object pose estimation and segmentation in industrial bin-picking, including synthetic and real data with detailed annotations, enabling improved learning-based approaches.

Contribution

The paper presents the first publicly available, extensively annotated dataset for 6D pose estimation and segmentation in bin-picking, combining synthetic and real-world scenes.

Findings

Dataset includes diverse synthetic and real scenes.

Provides precise annotations for pose, visibility, and segmentation.

Facilitates development of learning-based bin-picking methods.

Abstract

In this paper, we introduce a new public dataset for 6D object pose estimation and instance segmentation for industrial bin-picking. The dataset comprises both synthetic and real-world scenes. For both, point clouds, depth images, and annotations comprising the 6D pose (position and orientation), a visibility score, and a segmentation mask for each object are provided. Along with the raw data, a method for precisely annotating real-world scenes is proposed. To the best of our knowledge, this is the first public dataset for 6D object pose estimation and instance segmentation for bin-picking containing sufficiently annotated data for learning-based approaches. Furthermore, it is one of the largest public datasets for object pose estimation in general. The dataset is publicly available at http://www.bin-picking.ai/en/dataset.html.