GraspNet: A Large-Scale Clustered and Densely Annotated Dataset for Object Grasping

TL;DR

GraspNet introduces a large-scale, densely annotated RGBD dataset with an evaluation system for object grasping, addressing data scarcity and benchmarking issues in clustered scenes.

Contribution

It provides a comprehensive dataset with over 87,000 images and 370 million grasp poses, along with an evaluation system that simplifies success assessment without exhaustive labeling.

Findings

Dataset aligns well with real-world experiments

Evaluation system accurately assesses grasp success

Large-scale data improves training for grasp detection models

Abstract

Object grasping is critical for many applications, which is also a challenging computer vision problem. However, for the clustered scene, current researches suffer from the problems of insufficient training data and the lacking of evaluation benchmarks. In this work, we contribute a large-scale grasp pose detection dataset with a unified evaluation system. Our dataset contains 87,040 RGBD images with over 370 million grasp poses. Meanwhile, our evaluation system directly reports whether a grasping is successful or not by analytic computation, which is able to evaluate any kind of grasp poses without exhausted labeling pose ground-truth. We conduct extensive experiments to show that our dataset and evaluation system can align well with real-world experiments. Our dataset, source code and models will be made publicly available.