Extracting essential local object characteristics for 3D object categorization

Most object classes share a considerable amount of local appearance and often only a small number of features are discriminative. The traditional approach to represent an object is based on a summarization of the local characteristics by counting the number of feature occurrences. In this paper we propose the use of a recently developed technique for summarizations that, rather than looking into the quantity of features, encodes their quality to learn a description of an object. Our approach is based on extracting and aggregating only the essential characteristics of an object class for a task. We show how the proposed method significantly improves on previous work in 3D object categorization. We discuss the benefits of the method in other scenarios such as robot grasping. We provide extensive quantitative and qualitative experiments comparing our approach to the state of the art to justify the described approach.

Authors

• Marianna Madry
• Heydar Maboudi Afkham
• Carl Henrik Ek
• Stefan Carlsson
• Danica Kragic

Keywords

• Feature extraction
• Three-dimensional displays
• Histograms
• Databases
• Animals
• Robustness
• System-on-chip
• Local Features
• Essential Features
• Object Features
• Essential Objective
• Feature Classification
• Object Classification
• Qualitative Results
• Point Cloud
• Hyperplane
• Global Structure
• Dirac Delta
• Quantitative Description
• Class Position
• Objective Structured
• Large Separation
• Local Descriptors
• Final Representation
• Object Scale
• Global Descriptors
• Object Pose
• Object Instances
• Toilet Paper
• Object Point Cloud
• Histogram Features
• Incompleteness Of Data
• Fast Feature
• Sparse Representation
• Scale Variation
• Single Object