Attention-based active 3D point cloud segmentation

In this paper we present a framework for the segmentation of multiple objects from a 3D point cloud. We extend traditional image segmentation techniques into a full 3D representation. The proposed technique relies on a state-of-the-art min-cut framework to perform a fully 3D global multi-class labeling in a principled manner. Thereby, we extend our previous work in which a single object was actively segmented from the background. We also examine several seeding methods to bootstrap the graphical model-based energy minimization and these methods are compared over challenging scenes. All results are generated on real-world data gathered with an active vision robotic head. We present quantitive results over aggregate sets as well as visual results on specific examples.

Authors

• Matthew Johnson-Roberson
• Jeannette Bohg
• Mårten Björkman
• Danica Kragic

Keywords

• Clouds
• Image segmentation
• Three dimensional displays
• Image color analysis
• Humans
• Minimization
• Labeling
• Point Cloud
• 3D Point Cloud
• Activation Segment
• Point Cloud Segmentation
• Energy Minimization
• Multi-label
• Multiple Objects
• Single Object
• Object Segmentation
• 3D Representation
• Segmentation Techniques
• Segmentation Framework
• Traditional Segmentation
• Random Fields
• Energy Function
• Precision And Recall
• Graphical Model
• Gaussian Mixture Model
• Central Objective
• Graphical Framework
• Markov Random Field
• Foreground Objects
• Background Objects
• Conditional Random Field
• Color Model
• Segmentation Procedure
• Input Point Cloud