Results 1  10
of
30
On the Wiberg algorithm for matrix factorization in the presence of missing components
 INTERNATIONAL JOURNAL OF COMPUTER VISION
, 2006
"... This paper considers the problem of factorizing a matrix with missing components into a product of two smaller matrices, also known as principal component analysis with missing data (PCAMD). The Wiberg algorithm is a numerical algorithm developed for the problem in the community of applied mathemati ..."
Abstract

Cited by 26 (1 self)
 Add to MetaCart
This paper considers the problem of factorizing a matrix with missing components into a product of two smaller matrices, also known as principal component analysis with missing data (PCAMD). The Wiberg algorithm is a numerical algorithm developed for the problem in the community of applied mathematics. We argue that the algorithm has not been correctly understood in the computer vision community. Although there are many studies in our community, almost every one of which refers to the Wiberg study, as far as we know, there is no literature in which the performance of the Wiberg algorithm is investigated or the detail of the algorithm is presented. In this paper, we present derivation of the algorithm along with a problem in its implementation that needs to be carefully considered, and then examine its performance. The experimental results demonstrate that the Wiberg algorithm shows a considerably good performance, which should contradict the conventional view in our community, namely that minimizationbased algorithms tend to fail to converge to a global minimum relatively frequently. The performance of the Wiberg algorithm is such that even starting with random initial values, it converges in most cases to a correct solution, even when the matrix has many missing components and the data are contaminated with very strong noise. Our conclusion is that the Wiberg algorithm can also be used as a standard algorithm for the problems of computer vision. 3 1
Yet Another AppearanceBased Method for Pose Estimation Based on a Linear Model
"... This paper explores the possibility of a linear model as a solution to the problem of appearancebased pose estimation. The parametric eigenspace method (or its extensions that are based on correlation between images) has been widely used and yields successful results for the pose estimation problem ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
This paper explores the possibility of a linear model as a solution to the problem of appearancebased pose estimation. The parametric eigenspace method (or its extensions that are based on correlation between images) has been widely used and yields successful results for the pose estimation problem. On the other hand, the method has some problems. One is that large computational cost and storage space are required. Another is that small changes in appearance can be discarded even if it is related to changes of parameters to be estimated. Based on these, another appearancebased method for estimating pose of an object using a linear model is examined. Experimental results are not superior to the eigenspace method in terms of estimation accuracy. However, it has several advantages to the parametric eigenspace method in terms of storage space and computational cost, and has several features that can be advantages to the eigenspace method. 1
unknown title
"... Detailed derivation of the MF and BP algorithms We show here detailed derivation of the new MF and BP algorithms. In our main paper, we present simplified derivation which needs an assumption that the number S i of mixtures is the same for all sites, i.e., S i = S for any i. In what follows, we pre ..."
Abstract
 Add to MetaCart
Detailed derivation of the MF and BP algorithms We show here detailed derivation of the new MF and BP algorithms. In our main paper, we present simplified derivation which needs an assumption that the number S i of mixtures is the same for all sites, i.e., S i = S for any i. In what follows, we present complete derivation of the two algorithms which do not require this assumption; several equations that are omitted in the main paper are also given. Our derivation follows that of conventional MF and BP algorithms in [2]. Derivation of the new MF algorithm As mentioned in our main paper, MF and BP algorithms find P that minimizes the following free energy: F[P] = 〈E〉P − S [P], (30) where the first term is the expectation defined as 〈E〉P =∫ P(x)E(x)dx, and the second term is the entropy of P, i.e., S [P] = − ∫ P(x) ln P(x)dx. The derivation of MF algorithms start with assuming that the variable of each site i is independent of that of any other site: P(x) ≡
Projectorscreencamera system: Theory and algorithm for screentocamera homography estimation
 in Proc. International Conference on Computer Vision
, 2003
"... This paper deals with the autocalibration of a system that consists of a planar screen, multiple projectors, and a camera. In the system, either multiple projectors or a single moving projector projects patterns on a screen while a stationary camera placed in front of the screen takes images of the ..."
Abstract

Cited by 11 (0 self)
 Add to MetaCart
This paper deals with the autocalibration of a system that consists of a planar screen, multiple projectors, and a camera. In the system, either multiple projectors or a single moving projector projects patterns on a screen while a stationary camera placed in front of the screen takes images of the patterns. We treat the case in which the patterns that the projectors project toward space are assumed to be known (i.e., the projectors are calibrated), whereas poses of the projectors are unknown. Under these conditions, we consider the problem of estimating screentocamera homography from the images alone. This is intended for cases where there is no clue on the screen surface that enables direct estimation of the screentocamera homography. One application is a 6DOF input device; poses of a multibeam projector freely moving in space are computed from the images of beam spots on the screen. The primary contribution of the paper is theoretical results on the uniqueness of solutions and a noniterative algorithm for the problem. The effectiveness of the method is shown by experimental results on synthetic as well as on real images. 1.
1 23 Recovering Camera Motion from Image Sequence Based on Registration of Silhouette Cones: Shape from Silhouette Using a Mobile Camera with a Gyro Sensor
"... A method for reconstructing shape of an object from its silhouette using a mobile camera to which a gyro sensor is attached is proposed. In order to determine unknown camera positions at which images are taken, the pose information of the camera derived from the attached gyro sensor as well as silho ..."
Abstract
 Add to MetaCart
A method for reconstructing shape of an object from its silhouette using a mobile camera to which a gyro sensor is attached is proposed. In order to determine unknown camera positions at which images are taken, the pose information of the camera derived from the attached gyro sensor as well as silhouette of the object are used. An algorithm for computing the camera positions by an iterative process of registering silhouettes associated with viewpoints is shown. After the computation of the camera positions, the object shape is reconstructed based on the usual shape from silhouette algorithm. The camera is mobile, and thus it can take silhouette images of the object from arbitrary directions. This enables us to avoid incorrect shape reconstruction due to restricted viewing angles, which often occurs in conventional shape from silhouette methods. Several experimental results are shown. 1
Motion Parameter Estimation from Optical Flow without Nuisance Parameters
, 2003
"... Many kinds of computer vision problems can be formalized as statistical estimation problems with nuisance parameters. In the past, such problems have been solved without making any distinction between the nuisance parameters and structural ones. However, a theory of statistics suggests that eliminat ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
Many kinds of computer vision problems can be formalized as statistical estimation problems with nuisance parameters. In the past, such problems have been solved without making any distinction between the nuisance parameters and structural ones. However, a theory of statistics suggests that eliminating the nuisance parameters by assuming a probability distribution on them improves estimation accuracy of the structural parameters. In this paper, we apply this strategy to problem of estimating motion parameters from optical flow, which is a typical computer vision problem, and compare the estimation accuracy with that obtained by the conventional estimation method.
3Dimensional Tracking of Multiple Object Motions from Multiview Images by Mixedstate CONDENSATION Algorithm
"... In this paper, we propose an efficient method for tracking multiple objects, which deals with occlusion situations in an image. We use Shape from silhouettes method to reconstruct objects 3D position from images taken from different directions. Then we track the objects' reconstructed 3D positi ..."
Abstract
 Add to MetaCart
In this paper, we propose an efficient method for tracking multiple objects, which deals with occlusion situations in an image. We use Shape from silhouettes method to reconstruct objects 3D position from images taken from different directions. Then we track the objects' reconstructed 3D positions by using mixedstate CONDENSATION algorithm. The experimental results show the robust tracking of multiple objects. 1 Introduction Figure 1: Understanding 3D dynamic erivironment In this paper, we treat the situation shown as Fig.1, where multiple cameras observe a scene and track moving multiple objects in the scene. We achieve robust identification of multiple moving objects by fusing multiple
A PhysicsBased Imaging Model of Scanning Electron Microscopes
"... This paper discusses a physicsbased imaging model of scanning electron microscopes (SEM). The purpose is to accurately examine the imaging process of a SEM, which has to be necessary to realize novel applications of the SEM images, such as 3D shape reconstruction from image brightness. Its brightne ..."
Abstract
 Add to MetaCart
This paper discusses a physicsbased imaging model of scanning electron microscopes (SEM). The purpose is to accurately examine the imaging process of a SEM, which has to be necessary to realize novel applications of the SEM images, such as 3D shape reconstruction from image brightness. Its brightness is determined by the total energy of the secondary electrons derived by the incidence of accelerated electron beam to a surface point and then captured by the SEM detector. There are several simple imaging models. But, they are pointed out that the actual brightness cannot be dealt with. We then develop a better imaging model that precisely describes the physical process of the emergence of the secondary electron, their reflections and detections. 1
Hand Gesture Recognition using Histogram of Oriented Gradients and Partial Least Squares Regression
"... In this paper we propose a realtime hand gesture recognition system that employs the techniques developed for pedestrian detection to recognize a small vocabulary of human hand gestures. Our feature set comprises of grids of Histogram of Oriented Gradient (HOG) descriptors, with fine orientation bi ..."
Abstract
 Add to MetaCart
In this paper we propose a realtime hand gesture recognition system that employs the techniques developed for pedestrian detection to recognize a small vocabulary of human hand gestures. Our feature set comprises of grids of Histogram of Oriented Gradient (HOG) descriptors, with fine orientation binning and multilevel spatial binning for getting descriptors at the small as well as large scale. The overlapping descriptor blocks, which are contrast normalized to handle illumination changes, have a high degree of multicollinearity, resulting in a feature set of high dimensionality (more than 8000 dimensions), rendering it unsuitable for classification using the classical machine learning algorithms. Thus, we employ Partial Least Squares (PLS) regression as a ‘class aware’ method of dimensionality reduction, to project the feature vectors on to a lower dimensional space of 10 dimensions. We examine the results obtained by PLS as well as Principal Component Analysis (PCA) which show, that PLS outperforms PCA, and gives a better projection which preserves significant discriminative information. 1.
On Uniqueness of Solutions of the ThreeLightSource Photometric Stereo: Conditions on Illumination Configuration
, 2000
"... This paper is concerned with photometric methods using three images with different lighting direction to obtain shape information of an object. Such methods are based on the photometric equation that relates the normal of the object surface to the triplet of the image brightness. This paper discuss ..."
Abstract
 Add to MetaCart
This paper is concerned with photometric methods using three images with different lighting direction to obtain shape information of an object. Such methods are based on the photometric equation that relates the normal of the object surface to the triplet of the image brightness. This paper discusses the issue of whether the surface normal and the orientation of the 3vector formed by the image brightness triplet is onetoone in the equation. Several types of photometric methods require this relation to be onetoone. We mainly consider the case where the reflectance map is an increasing function of the angle between the surface normal and the illuminant direction. We first point out that even in this simple case, it is possible that the relation is not onetoone. Then we derive several sufficient conditions on the reflectance as well as the illumination configuration for the onetoone relation. c ° 2001 Academic Press 1.
Results 1  10
of
30