Title :
Trainable attention model based vergence control for active stereo vision system
Author :
Choi, Sang-Bok ; Jung, Bum-Soo ; Ban, Sang-Woo ; Lee, Minho
Author_Institution :
Dept. of Sensor Eng., Kyungpook Nat. Univ., Taegu, South Korea
Abstract :
We propose a new human-like vergence control method for an active stereo vision system. The proposed system uses a trainable selective attention model to localize an interesting object area in each camera. The selected object area in the master camera is compared with that in the slave camera to identify whether the two cameras find the same landmark. If the left and right cameras successfully find the same landmark, the implemented active vision system, with two cameras, focuses on the landmark. Using the motor encoder information, we can detect the depth information automatically. Computer simulation and experimental results show that the proposed vergence control method is very effective in implementing the human-like active stereo vision system.
Keywords :
ART neural nets; active vision; distance measurement; fuzzy neural nets; self-focusing; stereo image processing; biologically motivated vergence control method; depth information detection; focusing; fuzzy ART neural network; human-like active stereo vision system; image landmark determination; interesting object area localization; motor encoder information; trainable selective attention model; twin-camera system; vision system vergence control; Biological control systems; Biological system modeling; Brain modeling; Cameras; Control systems; Humans; Layout; Machine vision; Samarium; Stereo vision;
Conference_Titel :
Intelligent Sensors, Sensor Networks and Information Processing Conference, 2004. Proceedings of the 2004
Print_ISBN :
0-7803-8894-1
DOI :
10.1109/ISSNIP.2004.1417515