We present an active vision algorithm for computing the orientation and position of a locally planar object, onto which is cast a shadow of the edge of a half-plane at an unknown location. This algorithm utilises active position control of a point light source, and employs a Kalman filter to perform temporal integration of measurements. The light source position is adjusted after each measurement so as to reduce the trace of the expected state estimate error covariance matrix for the next measurement. We demonstrate the active shape-from-shadows algorithm using a real robotic system.
- Active vision
- Kalman filter
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition
- Artificial Intelligence