Results 1 
7 of
7
Merging and Splitting Eigenspace Models
, 2000
"... We present new deterministic methods that given two eigenspace models, each representing a set of ndimensional observations will: (1) merge the models to yield a representation of the union of the sets; (2) split one model from another to represent the difference between the sets; as this is done, ..."
Abstract

Cited by 57 (0 self)
 Add to MetaCart
(Show Context)
We present new deterministic methods that given two eigenspace models, each representing a set of ndimensional observations will: (1) merge the models to yield a representation of the union of the sets; (2) split one model from another to represent the difference between the sets; as this is done, we accurately keep track of the mean.
Eigenspace Updating for NonStationary Process and Its Application to Face Recognition
, 2002
"... In this paper, we introduce a novel approach to modeling nonstationary random processes. Given a set of training samples sequentially, we can iteratively update an eigenspace to manifest the current statistics provided by each new sample. The updated eigenspace is derived more from recent samples a ..."
Abstract

Cited by 14 (3 self)
 Add to MetaCart
(Show Context)
In this paper, we introduce a novel approach to modeling nonstationary random processes. Given a set of training samples sequentially, we can iteratively update an eigenspace to manifest the current statistics provided by each new sample. The updated eigenspace is derived more from recent samples and less from older samples, controlled by a number of decay parameters. Extensive study has been performed on how to choose these decay parameters. Other existing eigenspace updating algorithms can be regarded as special cases of our algorithm. We show the effectiveness of the proposed algorithm with both synthetic data and practical applications for face recognition. Significant improvements have been observed in recognizing face images with different variations, such as pose, expression and illumination variations. We also expect the proposed algorithm to have other applications in active recognition and modeling.
ScaleInvariant Image Recognition Based On Higher Order Autocorrelation Features
 Pattern Recognition
, 1996
"... We propose a framework and a complete implementation of a translation and scale invariant image recognition system for natural indoor scenes. The system employs higher order autocorrelation features of scale space data which permit linear classification. An optimal linear classification method is pr ..."
Abstract

Cited by 12 (1 self)
 Add to MetaCart
We propose a framework and a complete implementation of a translation and scale invariant image recognition system for natural indoor scenes. The system employs higher order autocorrelation features of scale space data which permit linear classification. An optimal linear classification method is presented, which is able to cope with a large number of classes represented by many, as well as very few samples. In the course of the analysis of our system, we examine which numerical methods for feature transformation and classification show sufficient stability to fulfill these demands. The implementation has been extensively tested. We present the results of our own application and several classification benchmarks. Image recognition Face recognition Scale invariancy Scale space Higher order autocorrelation Optimal linear classification 1. INTRODUCTION The task of visual recognition which was defined by Marr (1) with the question: "What objects are where in the environment?" is still ...
Tracking a Few Extreme Singular Values and Vectors in Signal Processing
"... In various applications, it is necessary to keep track of a lowrank approximation of a covariance matrix, R(t), slowly varying with time. It is convenient to track the left singular vectors associated with the largest singular values of the triangular factor, L(t), of its Cholesky factorization. Th ..."
Abstract
 Add to MetaCart
In various applications, it is necessary to keep track of a lowrank approximation of a covariance matrix, R(t), slowly varying with time. It is convenient to track the left singular vectors associated with the largest singular values of the triangular factor, L(t), of its Cholesky factorization. These algorithms are referred to as “squareroot.” The drawback of the Eigenvalue Decomposition (€VD) or the Singular Value Decomposition (SVD) is usually the volume of the computations. Various numerical methods carrying out this task are surveyed in this paper, and we show why this admittedly heavy computational burden is questionable in numerous situations and should be revised. Indeed, the complexity per eigenpair is generally a quadratic function of the problem size, but there exist faster algorithms whose complexity is linear. Finally, in order to make a choice among the large and fuzzy set of available techniques, comparisons are made based on computer simulations in a relevant signal processing context. I.
A MultiLevel Approach for Video Temporal Segmentation based on Adaptive Examples
, 2006
"... ii Over the past decade, various methods for video shot boundary detection and classification have been proposed and implemented by numerous researchers. Nonetheless, many of these techniques are specific to the type of transition or they are more complicated than necessary. Our research is develope ..."
Abstract
 Add to MetaCart
(Show Context)
ii Over the past decade, various methods for video shot boundary detection and classification have been proposed and implemented by numerous researchers. Nonetheless, many of these techniques are specific to the type of transition or they are more complicated than necessary. Our research is developed around the idea of creating a simple, realtime and general algorithm which can be used to detect both effects as well as transitions within a videostream segment. We have implemented several novel methods which have led to such a system. Adaptive examples, use of no thresholds, parameter free algorithms, extremely sensitive change detectors, parallel analyzers and uncertainty groups are among these methods.
Eigenspace Updating With Temporal Weighting for Face Analysis
, 2002
"... In this paper, we introduce a novel approach to eigenspace updating with temporal weighting. Given a set of training samples sequentially, we can iteratively update the eigenspace to manifest the current statistics brought by each new sample. The updated eigenspace can model more the most recent sam ..."
Abstract
 Add to MetaCart
In this paper, we introduce a novel approach to eigenspace updating with temporal weighting. Given a set of training samples sequentially, we can iteratively update the eigenspace to manifest the current statistics brought by each new sample. The updated eigenspace can model more the most recent samples and less and less for older samples. Thus it captures the nonstationary characteristics' of the samples. We show the effectiveness of our algorithm with practical applications on detection of facial expression changes and face recognition.
time
"... Microsensors and image processing for single oocyte qualification: towards multiparametric determination of the best ..."
Abstract
 Add to MetaCart
(Show Context)
Microsensors and image processing for single oocyte qualification: towards multiparametric determination of the best