Hopfield network for stereo vision correspondence
- 1 January 1992
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 3 (1) , 5-13
- https://doi.org/10.1109/72.105413
Abstract
An optimization approach is used to solve the correspondence problem for a set of features extracted from a pair of stereo images. A cost function is defined to represent the constraints on the solution, which is then mapped onto a two-dimensional Hopfield neural network for minimization. Each neuron in the network represents a possible match between a feature in the left image and one in the right image. Correspondence is achieved by initializing (exciting) each neuron that represents a possible match and then allowing the network to settle down into a stable state. The network uses the initial inputs and the compatibility measures between the matched points to find a stable state.Keywords
This publication has 22 references indexed in Scilit:
- A hierarchical multiple-view approach to three-dimensional object recognitionIEEE Transactions on Neural Networks, 1991
- Stereo vision correspondence using a multichannel graph matching techniqueImage and Vision Computing, 1989
- Integration of stereo vision and optical flow by using an energy-minimization approachJournal of the Optical Society of America A, 1989
- Structure from stereo-a reviewIEEE Transactions on Systems, Man, and Cybernetics, 1989
- Image restoration using a neural networkIEEE Transactions on Acoustics, Speech, and Signal Processing, 1988
- Pattern recognition by labeled graph matchingNeural Networks, 1988
- Ill-posed problems in early visionProceedings of the IEEE, 1988
- Segment-based stereo matchingComputer Vision, Graphics, and Image Processing, 1985
- Optimization by Simulated AnnealingScience, 1983
- Binocular Vision: A Physical and a Neural TheoryThe American Journal of Psychology, 1970