This contribution proposes a method to generate object hypotheses from stereo obstacle detection and image motion. Our algorithm is a general approach since it does not require any a priori information about the shape of the observed objects but relies on the basic assumption that the objects are rigid. The algorithm has two processing stages: First, obstacles are detected using stereo vision. Second, each obstacle is segmented into clusters of consistent motion in 3D space. The clustering process explicitly accounts for measurement uncertainties of stereo disparity and 2d motion. Our system may serve as a general feature for higher-level object detection and classification.
Fast Object Hypotheses Generation Using 3D Position and 3D Motion
2005-01-01
735163 byte
Conference paper
Electronic Resource
English
About Robust Hypotheses Generation and Object Validation in Traffic Scenes
British Library Conference Proceedings | 2007
|OBJECT POSITION MEASUREMENT WITH AUTOMOTIVE CAMERA USING VEHICLE MOTION DATA
European Patent Office | 2016
|Vulnerability classification of generic object hypotheses using a visual words approach
Automotive engineering | 2014
|OBJECT POSITION MEASUREMENT WITH AUTOMOTIVE CAMERA USING VEHICLE MOTION DATA
European Patent Office | 2018
|