Edge Grasp Network: A Graph-Based SE(3)-invariant Approach to Grasp Detection

Given point cloud input, the problem of 6-DoF grasp pose detection is to identify a set of hand poses in SE(3) from which an object can be successfully grasped. This important problem has many practical applications. Here we propose a novel method and neural network model that enables better grasp success rates relative to what is available in the literature. The method takes standard point cloud data as input and works well with single-view point clouds observed from arbitrary viewing directions. Videos and code are available at https://haojhuang.github.io/edge_grasp_page/.

Authors

• Haojie Huang 0001
• Dian Wang 0001
• Xupeng Zhu
• Robin Walters 0001
• Robert Platt 0001

Keywords

• Point cloud compression
• Codes
• Automation
• Image edge detection
• Neural networks
• Grasping
• Feature extraction
• Artificial Neural Network
• Neural Model
• Point Cloud
• Input Point Cloud
• Percentage Points
• Data Augmentation
• Global Features
• Feature Points
• Depth Images
• Equivalency
• Max-pooling Layer
• Vertices
• Graph Neural Networks
• Edge Features
• Rotation Invariance
• Strong Baseline
• Sampling-based Methods