Open Access
Hand-printed digit recognition using deformable models
Christopher Williams,Michael Revow,Geoffrey E. Hinton +2 more
- pp 127-147
Reads0
Chats0
TLDR
This work uses an elastic matching algorithm to minimize an energy function that includes both the deformation energy of the digit model and the log probability that the model would generate the inked pixels in the image.Abstract:
Deformable models are an attractive way for characterizing handwritten digits since they have relatively few parameters, are able to capture many topological variations, and incorporate much prior knowledge. We have described a system [8] that uses learned digit models consisting of splines whose shape is governed by a small number of control points. Images can be classi ed by separately tting each digit model to the image, and using a simple neural network to decide which model ts best. We use an elastic matching algorithm to minimize an energy function that includes both the deformation energy of the digit model and the log probability that the model would generate the inked pixels in the image. The use of multiple models for each digit can characterize the population of handwritten digits better. We show how multiple models may be used without increasing the time required for elastic matching.read more
Citations
More filters
Journal ArticleDOI
GTM: the generative topographic mapping
TL;DR: A form of nonlinear latent variable model called the generative topographic mapping, for which the parameters of the model can be determined using the expectation-maximization algorithm, is introduced.
Journal ArticleDOI
Using generative models for handwritten digit recognition
TL;DR: A method of recognizing handwritten digits by fitting generative models that are built from deformable B-splines with Gaussian "ink generators" spaced along the length of the spline using a novel elastic matching procedure based on the expectation maximization algorithm.
Journal ArticleDOI
Developments of the generative topographic mapping
TL;DR: Several extensions of the generative topographic mapping model are reported, including an incremental version of the EM algorithm for estimating the model parameters, the use of local subspace models, extensions to mixed discrete and continuous data, semi-linear models which permit the useof high-dimensional manifolds whilst avoiding computational intractability, Bayesian inference applied to hyper-parameters, and an alternative framework for the GTM based on Gaussian processes.
Proceedings ArticleDOI
A Generic System For Classifying Variable Objects Using Flexible Template Matching
TL;DR: A technique for classifying variable objects using flexible template models is described and is recognised as the input, plant seeds, handprinted characters and human faces.
Dissertation
Recognition of handwritten numerals using elastic matching
TL;DR: Recognition of Handwritten Numerals Using Elastic Matching Patrice Scattolin Elastic matching has been used for the recognition of handwritten characters for two decades but is usually only used for writer-dependent systems with on-line data.
References
More filters
Journal ArticleDOI
Maximum likelihood from incomplete data via the EM algorithm
Book ChapterDOI
Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition
TL;DR: In this article, the outputs of the network are treated as probabilities of alternatives (e.g. pattern classes), conditioned on the inputs, and two modifications are proposed: probability scoring, which is an alternative to squared error minimisation, and a normalised exponential (softmax) multi-input generalisation of the logistic nonlinearity.
Journal ArticleDOI
An analogue approach to the travelling salesman problem using an elastic net method
TL;DR: This work describes how a parallel analogue algorithm, derived from a formal model for the establishment of topographically ordered projections in the brain, can be applied to the travelling salesman problem, and produces shorter tour lengths than another recent parallel analogue algorithms.
Proceedings Article
Efficient Pattern Recognition Using a New Transformation Distance
TL;DR: A new distance measure which can be made locally invariant to any set of transformations of the input and can be computed efficiently is proposed.