Diffusion Suction Grasping with Large-Scale Parcel Dataset

2025-02-11

AI Summary

Suction-based grasping of densely stacked parcels faces two key bottlenecks: the absence of a dedicated, high-quality dataset and poor generalization to diverse parcel sizes, geometries, and surface textures. Method: We introduce Parcel-Suction-Dataset—the first large-scale synthetic dataset containing 25K scenes and 410M physically feasible suction poses—and propose Diffusion-Suction, a novel framework featuring (i) a geometry-aware sampling algorithm that generates suction poses compatible with material properties and physical constraints, and (ii) the first application of denoising diffusion probabilistic models (DDPMs) to suction grasping, enabling iterative, vision-guided generation of spatial point-level grasp confidence maps. Contribution/Results: Our method achieves state-of-the-art performance on both Parcel-Suction-Dataset and SuctionNet-1Billion, significantly improving grasp success rates in complex, cluttered parcel scenarios and demonstrating superior cross-domain generalization.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: Manipulation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

While recent advances in object suction grasping have shown remarkable progress, significant challenges persist particularly in cluttered and complex parcel handling scenarios. Two fundamental limitations hinder current approaches: (1) the lack of a comprehensive suction grasp dataset tailored for parcel manipulation tasks, and (2) insufficient adaptability to diverse object characteristics including size variations, geometric complexity, and textural diversity. To address these challenges, we present Parcel-Suction-Dataset, a large-scale synthetic dataset containing 25 thousand cluttered scenes with 410 million precision-annotated suction grasp poses. This dataset is generated through our novel geometric sampling algorithm that enables efficient generation of optimal suction grasps incorporating both physical constraints and material properties. We further propose Diffusion-Suction, an innovative framework that reformulates suction grasp prediction as a conditional generation task through denoising diffusion probabilistic models. Our method iteratively refines random noise into suction grasp score maps through visual-conditioned guidance from point cloud observations, effectively learning spatial point-wise affordances from our synthetic dataset. Extensive experiments demonstrate that the simple yet efficient Diffusion-Suction achieves new state-of-the-art performance compared to previous models on both Parcel-Suction-Dataset and the public SuctionNet-1Billion benchmark.

Problem

• Large-scale parcel handling challenges
• Insufficient adaptability to object diversity
• Lack of comprehensive suction grasp dataset

Innovation

• Large-scale synthetic dataset creation
• Geometric sampling algorithm
• Denoising diffusion probabilistic models