Title :
A neural architecture for expectation driven detection of moving objects during egomotion
Abstract :
We present an architecture functionally inspired by biology for drawing selective attention to visually unexpected and therefore conspicuous events or image changes. This ability is investigated by means of a mobile robot scenario, whereby the observing system is moving around and detects other dynamic objects within the scene based on vision only. After detection of a dynamic object our system is able to track it or to realize a higher behavior like hunting or escaping
Keywords :
image sequences; mobile robots; neural net architecture; object detection; robot vision; conspicuous events; dynamic objects; egomotion; escaping; expectation driven detection; higher behavior; hunting; image changes; mobile robot; moving objects; neural architecture; observing system; visually unexpected events; Cameras; Character generation; Image motion analysis; Layout; Mobile robots; Object detection; Predictive models; Robot vision systems; Sequences; Visual perception;
Conference_Titel :
Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7044-9
DOI :
10.1109/IJCNN.2001.938502
