A Probabilistic Approach to Simultaneous Segmentation, Object Recognition, 3D Localization, and Tracking Using Stereo

Vision systems for service robotics applications have to cope with varying environmental conditions, partial occlusions, complex backgrounds and a large number of distractors (clutter) present in the scene. This paper presents a new approach targeted at such application scenarios that combines segmentation, object recognition, 3D localization and tracking in a seamlessly integrated fashion. The unifying framework is the probabilistic representation of various aspects of the scene. Experiments indicate that this approach is viable and gives very satisfactory results.

• Type: Conference
• Year: 2001
• Where: DAGM
• Authors: Georg von Wichert
• Keywords: Complex Backgrounds, DAGM 2001, Image Processing, Partial Occlusions, Service Robotics Applications