Viewpoint detection models for sequential embodied object category recognition

This paper proposes a method for learning viewpoint detection models for object categories that facilitate sequential object category recognition and viewpoint planning. We have examined such models for several state-of-the-art object detection methods. Our learning procedure has been evaluated using an exhaustive multiview category database recently collected for multiview category recognition research. Our approach has been evaluated on a simulator that is based on real images that have previously been collected. Simulation results verify that our viewpoint planning approach requires fewer viewpoints for confident recognition. Finally, we illustrate the applicability of our method as a component of a completely autonomous visual recognition platform that has previously been demonstrated in an object category recognition competition.

Authors

• David Meger
• Ankur Gupta 0004
• James J. Little

Keywords

• Object detection
• Humans
• Robots
• Image databases
• Object recognition
• Bicycles
• Detectors
• Image recognition
• Thyristors
• Robotics and automation
• Category Recognition
• Object Category Recognition
• Recognition Site
• Object Detection Methods
• Recognizable
• General Class
• Image Object
• Detector Response
• Azimuth Angle
• Recognition System
• Visual Search
• Feature Matching
• Pose Estimation
• Category Labels
• Robot Motion
• Deformation Model
• Planning Algorithm
• Object Pose
• Instances Of Categories
• Object Labels
• Computer Vision Community
• Object Appearance
• Potential Objects
• Training Data
• Sequential Estimation
• Disambiguation
• Semantic