TY - GEN
T1 - Local reinforcement learning for object recognition
AU - Peng, Jing
AU - Bhanu, Bir
PY - 1998
Y1 - 1998
N2 - Current computer vision systems, whose basic methodology is open-loop or filter type, typically use image segmentation followed by object recognition algorithms. These systems are not robust for most real-world applications. In contrast, the system presented here achieves robust performance by using local reinforcement learning to induce a highly adaptive mapping from input images to segmentation strategies. This is accomplished by using the confidence level of model matching as reinforcement to drive learning. The system is verified through experiments on a large set of real images.
AB - Current computer vision systems, whose basic methodology is open-loop or filter type, typically use image segmentation followed by object recognition algorithms. These systems are not robust for most real-world applications. In contrast, the system presented here achieves robust performance by using local reinforcement learning to induce a highly adaptive mapping from input images to segmentation strategies. This is accomplished by using the confidence level of model matching as reinforcement to drive learning. The system is verified through experiments on a large set of real images.
UR - https://www.scopus.com/pages/publications/105035500019
U2 - 10.1109/ICPR.1998.711133
DO - 10.1109/ICPR.1998.711133
M3 - Conference contribution
AN - SCOPUS:105035500019
SN - 0818685123
SN - 9780818685125
T3 - Proceedings - International Conference on Pattern Recognition
SP - 272
EP - 274
BT - Proceedings - 14th International Conference on Pattern Recognition, ICPR 1998
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Conference on Pattern Recognition, ICPR 1998
Y2 - 16 August 1998 through 20 August 1998
ER -