Title :
Modified 3-D Hopfield neural network for gesture recognition
Author :
Huang, Chung-Lin ; Huang, Wen-Yi
Author_Institution :
Dept. of Electr. Eng., Nat. Tsing Hua Univ., Hsinchu, Taiwan
Abstract :
This paper presents a modified 3-D Hopfield neural network for gesture recognition. In the feature extraction phase, we use a model-based method to track hand motion and a Fourier descriptor to characterize hand figures. In the recognition phase, we propose a modified 3-D Hopfield neural network to perform graph matching between the input gesture model and the stored models to recognize the gesture
Keywords :
Hopfield neural nets; feature extraction; image sequences; motion estimation; Fourier descriptor; feature extraction; gesture recognition; graph matching; hand motion; model-based method; modified 3D Hopfield neural network; Biological system modeling; Data mining; Feature extraction; Hopfield neural networks; Image sequences; Impedance matching; Motion analysis; Shape; Tracking; Trajectory;
Conference_Titel :
Neural Networks,1997., International Conference on
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.614142
