Jacquard: A Large Scale Dataset for Robotic Grasp Detection

By Amaury Depierre, Emmanuel Dellandréa, Liming Chen. IEEE/RJS International Conference on Intelligent RObots and Systems, 2018

399 citations recorded by the source corpus.

Grasping skill is a major ability that a wide number of real-life applications require for robotisation. State-of-the-art robotic grasping methods perform prediction of object grasp locations based on deep neural networks. However, such networks require huge amount of labeled data for training making this approach often impracticable in robotics. In this paper, we propose a method to generate a large scale synthetic dataset with ground truth, which we refer to as the Jacquard grasping dataset. Jacquard is built on a subset of ShapeNet, a large CAD models dataset, and contains both RGB-D images and annotations of successful grasping positions based on grasp attempts performed in a simulated environment. We carried out experiments using an off-the-shelf CNN, with three different evaluation metrics, including real grasping robot trials. The results show that Jacquard enables much better generalization skills than a human labeled dataset thanks to its diversity of objects and grasping positions. For the purpose of reproducible research in robotics, we are releasing along with the Jacquard dataset a web interface for researchers to evaluate the successfulness of their grasping position detections using our dataset.

Subfields: Biomedical Engineering, Control and Systems Engineering, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/6272f7355e28148e576cad0a14e240c333299d78

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Control and Systems Engineering

In Human-Computer Interaction

Citations in the corpus

Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AI (2024, neutral): “2) Simulated Data: Data collection in real-world settings is resource-intensive and time-consuming, making simulation-based data collection an attractive alternative.”
GraspNet-1Billion: A Large-Scale Benchmark for General Object Grasping (2020, background): “To avoid such problem, some recent works explore using simulated environment [26, 9, 48, 28, 4] to anotate grasp poses.”
The CoSTAR Block Stacking Dataset: Learning with Workspace Constraints (2018, neutral): “Jacquard[33] 7 7 7 7 3 7 7 7 7 3 7 7 7 54,485 1 –”
AnyGrasp: Robust and Efficient Grasp Perception in Spatial and Temporal Domains (2022, background): “Besides providing data captured in real-world, another difference between GraspNet-1Billion [9] and other datasets [5], [46], [47], [56] is that the grasp pose annotation is much denser in [9].”
Learning robust, real-time, reactive robotic grasping (2019, methodology): “Both GG-CNNs outperform the original results from Depierre et al. (2018), which train an AlexNet network (approximately 60 million parameters) on the Jacquard dataset and evaluate under the same conditions, despite GG-CNN being orders of magnitude smaller.”
Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review (2020, methodology): “, 2020] began to conduct experiments on the Jacquard Grasp dataset [Depierre et al., 2018] since it has more images and the grasps are more diverse.”
Grasp'D: Differentiable Contact-rich Grasp Synthesis for Multi-fingered Hands (2022, methodology): “Although analytic metrics have been successfully applied to parallel-jaw gripper grasp synthesis (based on grasp wrench space analysis [27, 35, 63], robust grasp wrench space analysis [60, 86], or caging [62, 74]), more recent works [19, 24, 48, 66] have focused on simulation-based synthesis.”
Antipodal Robotic Grasping using Generative Residual Convolutional Neural Network (2019, methodology): “The first one is the Cornell grasp dataset [26], which is the most common grasping dataset used to benchmark results, and the second one is a more recent Jacquard grasping dataset [34], which is more than 50 times bigger the Cornell grasp dataset.”