Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
SuperGrasp addresses single-view robotic grasping by first matching input point clouds to a pre-computed superquadric primitive dataset to generate initial grasp candidates. Then, an end-to-end network (E-RNet) expands the grasp-aware region, using the initial grasp closure as a local anchor, to evaluate and refine these candidates. Experiments show SuperGrasp achieves stable grasping and generalization in simulation and real-world environments, outperforming existing methods.
SuperGrasp's two-stage approach to single-view grasping—using superquadric matching for initial candidates and a novel E-RNet for refinement—significantly boosts grasp stability and generalization, even with incomplete geometric information.
Robotic grasping from single-view observations remains a critical challenge in manipulation. Existing methods still struggle to generate stable and valid grasp poses when confronted with incomplete geometric information. To address these limitations, we propose SuperGrasp, a novel two-stage framework for single-view grasping with parallel-jaw grippers that decomposes the grasping process into initial grasp pose generation and subsequent grasp evaluation and refinement. In the first stage, we introduce a Similarity Matching Module that efficiently retrieves grasp candidates by matching the input single-view point cloud with a pre-computed primitive dataset based on superquadric coefficients. In the second stage, we propose E-RNet, an end-to-end network that expands the graspaware region and takes the initial grasp closure region as a local anchor region, enabling more accurate and reliable evaluation and refinement of grasp candidates. To enhance generalization, we construct a primitive dataset containing 1.5k primitives for similarity matching and collect a large-scale point cloud dataset with 100k stable grasp labels from 124 objects for network training. Extensive experiments in both simulation and realworld environments demonstrate that our method achieves stable grasping performance and strong generalization across varying scenes and novel objects.