Estimating the Pose and Motion of a Known Object for Real-Time Robotic Tracking

An approach for estimating the pose and motion of a known moving object in three dimensions from a sequence of monocular images is considered. The principle is to obtain initial estimates of the pose and motion parameters and to update them by using feature location measurements made from subsequent monocular ima.ge frames. The ultimate goal is to use the obtained estimates for controlling the movements of a robot arm.

• Type: Conference
• Year: 1990
• Where: MVA
• Authors: Olli Silvén
• Keywords: Initial Estimates, Monocular Images, MVA 1990, MVA 2007, Subsequent Monocular Ima.ge