SuperGrasp: Single-View Object Grasping via Superquadric Similarity Matching, Evaluation, and Refinement

TL;DR

SuperGrasp introduces a two-stage framework for single-view robotic grasping, combining primitive similarity matching with a neural network for accurate grasp evaluation and refinement, demonstrating robust performance in diverse scenarios.

Contribution

The paper presents a novel two-stage approach integrating primitive-based similarity matching and a specialized neural network for improved single-view grasping.

Findings

Achieves stable grasping in simulation and real-world tests.

Generalizes well to novel objects and cluttered scenes.

Constructs a large dataset for training and evaluation.

Abstract

Robotic grasping from single-view observations remains a critical challenge in manipulation. However, existing methods still struggle to generate reliable grasp candidates and stably evaluate grasp feasibility under incomplete geometric information. To address these limitations, we present SuperGrasp, a new two-stage framework for single-view parallel-jaw grasping. In the first stage, we introduce a Similarity Matching Module that efficiently retrieves valid and diverse grasp candidates by matching the input single-view point cloud with a precomputed primitive dataset based on superquadric coefficients. In the second stage, we propose E-RNet, an end-to-end network that expands the grasp-aware region and takes the initial grasp closure region as a local anchor region, capturing the contextual relationship between the local region and its surrounding spatial neighborhood, thereby enabling more accurate and reliable grasp evaluation and introducing small-range local refinement to improve grasp adaptability. To enhance generalization, we construct a primitive dataset containing 1.2k standard geometric primitives for similarity matching and collect a point cloud dataset of 100k samples from 124 objects, annotated with stable grasp labels for network training. Extensive experiments in both simulation and real-world environments demonstrate that our method achieves stable grasping performance and good generalization across novel objects and clutter scenes.