Robust Place Recognition using Local Appearance Based Methods

We present an approach to the automatic recognition of locations or landmarks using single camera images. Our approach is to learn visual features in the appearance domain that can be used to characterize an object or a location. These features are defined statistically and then are recognized using principal components in the frequency domain. We show that this technique can be used to recognize specific objects on varying backgrounds, as well as environmental features.

Authors

• Gregory Dudek
• Deeptiman Jugessur

Keywords

• Robustness
• Layout
• Robot sensing systems
• Mobile robots
• Image recognition
• Testing
• Smart cameras
• Frequency domain analysis
• Sensor phenomena and characterization
• Robot vision systems
• Place Recognition
• Robust Place Recognition
• Mobile Robot
• Partial Occlusion
• Scene Content
• Offline Learning
• Familiar Items
• View Of The Scene
• Diverse Backgrounds
• Training Phase
• Test Phase
• Singular Value
• Object Recognition
• Training Images
• Local Method
• Object Of Interest
• Entire Image
• Position Estimation
• Pose Estimation
• Attention Operation
• Eigenspace
• Rotation Invariance
• Principal Component Analysis Method
• Plane Rotation
• Degree Of Occlusion
• Polarization Imaging
• Neighboring Points
• Eigenvectors Of Matrix