Robust Statistics for 3D Object Tracking

This paper focuses on methods that enhance performance of a model based 3D object tracking system. Three statistical methods and an improved edge detector are discussed and compared. The evaluation is performed on a number of characteristic sequences incorporating shift, rotation, texture, weak illumination and occlusion. Considering the deviations of the pose parameters from ground truth, it is shown that improving the measurements' accuracy in the detection step yields better results than improving contaminated measurements with statistical means

Authors

• Peter Preisig
• Danica Kragic

Keywords

• Robustness
• Statistics
• Image edge detection
• Intelligent robots
• Pollution measurement
• Motion estimation
• Filtering
• Robot vision systems
• Intelligent systems
• Computational intelligence
• 3D Object Tracking
• Tracking System
• Edge Detection
• Pose Parameters
• Performance Of Algorithm
• Cost Function
• Image Plane
• Original Algorithm
• Consecutive Frames
• Pose Estimation
• Winsorized
• Search Region
• Robust Improvement
• Edges Of Objects
• Object Pose
• Front Edge
• Pose Changes
• Drop Ratio
• Background Texture
• Track Loss
• Back Edge
• Visible Edges