By Topic

Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.

27 June-2 July 2004

Go

Filter Results

Displaying Results 1 - 25 of 148
  • Computing depth under ambient illumination using multi-shuttered light

    Publication Year: 2004, Page(s):II-234 - II-241 Vol.2
    Cited by:  Papers (6)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (828 KB) | HTML iconHTML

    Range imaging has become a critical component of many computer vision applications. The quality of the depth data is of critical importance, but so is the need for speed. Shuttered light-pulse (SLP) imaging uses active illumination hardware to provide high quality depth maps at video frame rates. Unfortunately, current analytical models for deriving depth from SLP imagers are specific to the numbe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recovering shape and reflectance model of non-lambertian objects from multiple views

    Publication Year: 2004, Page(s):II-226 - II-233 Vol.2
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (874 KB) | HTML iconHTML

    This paper proposes an algorithm to simultaneously estimate both the 3D shape and parameters of a surface reflectance model from multiple views of an object made of a single material. The algorithm is based on a multiple view shape from shading method. A triangular mesh represents the shape of the object. The Phong reflectance model is used to model the surface reflectance. We iteratively find the... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Calibrating an air-ground control system from motion correspondences

    Publication Year: 2004, Page(s):II-218 - II-225 Vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (452 KB) | HTML iconHTML

    In this paper we consider the problem of controlling the motion of a vehicle moving on a ground plane based on aerial imagery. In the course of this work we propose a novel analysis of the relationship between the velocity of the vehicle on the ground plane and the velocity of its projection in the image. We show that this relationship provides information about a subset of the parameters of the h... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scale-invariant shape features for recognition of object categories

    Publication Year: 2004, Page(s):II-90 - II-96 Vol.2
    Cited by:  Papers (58)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1008 KB) | HTML iconHTML

    We introduce a new class of distinguished regions based on detecting the most salient convex local arrangements of contours in the image. The regions are used in a similar way to the local interest points extracted from gray-level images, but they capture shape rather than texture. Local convexity is characterized by measuring the extent to which the detected image contours support circle or arc-l... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 3D head tracking based on recognition and interpolation using a time-of-flight depth sensor

    Publication Year: 2004, Page(s):II-211 - II-217 Vol.2
    Cited by:  Papers (11)  |  Patents (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (382 KB) | HTML iconHTML

    This paper describes a head-tracking algorithm that is based on recognition and correlation-based weighted interpolation. The input is a sequence of 3D depth images generated by a novel time-of-flight depth sensor. These are processed to segment the background and foreground, and the latter is used as the input to the head tracking algorithm, which is composed of three major modules: First, a dept... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reconstructing open surfaces from unorganized data points

    Publication Year: 2004, Page(s):II-653 - II-660 Vol.2
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (896 KB) | HTML iconHTML

    In this paper a method for fitting open surfaces to an unorganized set of data points is presented using a level set representation of the surface. This is done by tracking a curve, representing the boundary, on the implicitly defined surface. This curve is given as the intersection of the level set describing the surface and an auxiliary level set. These two level sets are propagated using the sa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning distance functions for image retrieval

    Publication Year: 2004, Page(s):II-570 - II-577 Vol.2
    Cited by:  Papers (44)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (641 KB) | HTML iconHTML

    Image retrieval critically relies on the distance function used to compare a query image to images in the database. We suggest learning such distance functions by training binary classifiers with margins, where the classifiers are defined over the product space of pairs of images. The classifiers are trained to distinguish between pairs in which the images are from the same class and pairs, which ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parts-based 3D object classification

    Publication Year: 2004, Page(s):II-82 - II-89 Vol.2
    Cited by:  Papers (59)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (671 KB) | HTML iconHTML

    This paper presents a parts-based method for classifying scenes of 3D objects into a set of pre-determined object classes. Working at the part level, as opposed to the whole object level, enables a more flexible class representation and allows scenes in which the query object is significantly occluded to be classified. In our approach, parts are extracted from training objects and grouped into par... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Cyclic articulated human motion tracking by sequential ancestral simulation

    Publication Year: 2004, Page(s):II-45 - II-52 Vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (4877 KB) | HTML iconHTML

    Accurate tracking of cyclic human motion in video data helps in developing computer-aided applications such as gait analysis, visual surveillance, patient rehabilitation, etc. This paper presents a novel technique for tracking cyclic human motion based on decomposing complex cyclic motion into components and maintaining coupling between components. The decomposition reduces the dimensionality of t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Bridging the gaps between cameras

    Publication Year: 2004, Page(s):II-205 - II-210 Vol.2
    Cited by:  Papers (125)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (354 KB) | HTML iconHTML

    The paper investigates the unsupervised learning of a model of activity for a multi-camera surveillance network that can be created from a large set of observations. This enables the learning algorithm to establish links between camera views associated with an activity. The learning algorithm operates in a correspondence-free manner, exploiting the statistical consistency of the observation data. ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Elastic-string models for representation and analysis of planar shapes

    Publication Year: 2004, Page(s):II-10 - II-15 Vol.2
    Cited by:  Papers (15)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB) | HTML iconHTML

    We develop a new framework for the quantitative analysis of shapes of planar curves. Shapes are modeled on elastic strings that can be bent, stretched or compressed at different rates along the curve. Shapes are treated as elements of a space obtained as the quotient of an infinite-dimensional Riemannian manifold of elastic curves by the action of a reparameterization group. The Riemannian metric ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Random sampling based SVM for relevance feedback image retrieval

    Publication Year: 2004, Page(s):II-647 - II-652 Vol.2
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (322 KB) | HTML iconHTML

    Relevance feedback (RF) schemes based on support vector machine (SVM) have been widely used in content-based image retrieval. However, the performance of SVM based RF is often poor when the number of labeled positive feedback samples is small. This is mainly due to three reasons: (1) SVM classifier is unstable on small size training set; (2) SVM's optimal hyper-plane may be biased when the positiv... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dual-space linear discriminant analysis for face recognition

    Publication Year: 2004, Page(s):II-564 - II-569 Vol.2
    Cited by:  Papers (55)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (258 KB) | HTML iconHTML

    Linear discriminant analysis (LDA) is a popular feature extraction technique for face recognition. However, it often suffers from the small sample size problem when dealing with the high dimensional face data. Some approaches have been proposed to overcome this problem, but they are often unstable and have to discard some discriminative information. In this paper, a dual-space LDA approach for fac... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Searching the Web with mobile images for location recognition

    Publication Year: 2004, Page(s):II-76 - II-81 Vol.2
    Cited by:  Papers (44)  |  Patents (76)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (670 KB) | HTML iconHTML

    We describe an approach to recognizing location from mobile devices using image-based Web search. We demonstrate the usefulness of common image search metrics applied on images captured with a camera-equipped mobile device to find matching images on the World Wide Web or other general-purpose databases. Searching the entire Web can be computationally overwhelming, so we devise a hybrid image-and-k... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Is bottom-up attention useful for object recognition?

    Publication Year: 2004, Page(s):II-37 - II-44 Vol.2
    Cited by:  Papers (183)  |  Patents (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (740 KB) | HTML iconHTML

    A key problem in learning multiple objects from unlabeled images is that it is a priori impossible to tell which part of the image corresponds to each individual object, and which part is irrelevant clutter which is not associated to the objects. We investigate empirically to what extent pure bottom-up attention can extract useful information about the location, size and shape of objects from imag... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Uncontrolled modulation imaging

    Publication Year: 2004, Page(s):II-197 - II-204 Vol.2
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (581 KB) | HTML iconHTML

    To obtain high dynamic range or hyperspectral images, multiple frames of the same field of view are acquired while the imaging settings are modulated; images are taken at different exposures or through different wavelength bands. A major problem associated with such modulations has been the need for perfect synchronization between image acquisition and modulation control. In the past, this problem... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic view recognition in echocardiogram videos using parts-based representation

    Publication Year: 2004, Page(s):II-2 - II-9 Vol.2
    Cited by:  Papers (17)  |  Patents (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (490 KB) | HTML iconHTML

    Indexing echocardiogram videos at different levels of structure is essential for providing efficient access to their content for browsing and retrieval purposes. We present a novel approach for the automatic identification of the views of the heart from the content of the echocardiogram videos. In this approach the structure of the heart is represented by the constellation of its parts (chambers) ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Grouping with bias revisited

    Publication Year: 2004, Page(s):II-460 - II-465 Vol.2
    Cited by:  Papers (3)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (912 KB) | HTML iconHTML

    In this paper, we improve and tailor a recent statistical region merging approach to biased (partially supervised) grouping. The approach appears to be attractive both for its theoretical benefits and its experimental results, as light bias brings dramatic improvements over unbiased approaches on difficult pictures. Comparisons with another biased grouping algorithm display very favorable results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Color lines: image specific color representation

    Publication Year: 2004, Page(s):II-946 - II-953 Vol.2
    Cited by:  Papers (63)  |  Patents (24)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (941 KB) | HTML iconHTML

    The problem of deciding whether two pixels in an image have the same real world color is a fundamental problem in computer vision. Many color spaces are used in different applications for discriminating color from intensity to create an informative representation of color. The major drawback of all of these representations is that they assume no color distortion. In practice the colors of real wor... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning a restricted Bayesian network for object detection

    Publication Year: 2004, Page(s):II-639 - II-646 Vol.2
    Cited by:  Papers (37)  |  Patents (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (570 KB) | HTML iconHTML

    Many classes of images have the characteristics of sparse structuring of statistical dependency and the presence of conditional independencies among various groups of variables. Such characteristics make it possible to construct a powerful classifier by only representing the stronger direct dependencies among the variables. In particular, a Bayesian network compactly represents such structuring. H... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning classifiers from imbalanced data based on biased minimax probability machine

    Publication Year: 2004, Page(s):II-558 - II-563 Vol.2
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (368 KB) | HTML iconHTML

    We consider the problem of the binary classification on imbalanced data, in which nearly all the instances are labelled as one class, while far fewer instances are labelled as the other class, usually the more important class. Traditional machine learning methods seeking an accurate performance over a full range of instances are not suitable to deal with this problem, since they tend to classify a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Models of large population recognition performance

    Publication Year: 2004, Page(s):II-68 - II-75 Vol.2
    Cited by:  Papers (18)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (377 KB) | HTML iconHTML

    We present new binomial models of open- and closed-set identification recognition performance, giving formulae for identification and false match rates as functions of database size, match rank and operating threshold. We compare these with previously published models and with results from face recognition trials on populations of size 4 104. We note verification to be a special case of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feature-centric evaluation for efficient cascaded object detection

    Publication Year: 2004, Page(s):II-29 - II-36 Vol.2
    Cited by:  Papers (56)  |  Patents (25)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (605 KB) | HTML iconHTML

    We describe a cascaded method for object detection. This approach uses a novel organization of the first cascade stage called "feature-centric" evaluation which re-uses feature evaluations across multiple candidate windows. We minimize the cost of this evaluation through several simplifications: (1) localized lighting normalization, (2) representation of the classifier as an additive model and (3)... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient search of faces from complex line drawings

    Publication Year: 2004, Page(s):II-791 - II-796 Vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (303 KB) | HTML iconHTML

    Single 2D line drawing is a straightforward method to illustrate 3D objects. The faces of an object depicted by a line drawing give very useful information for the reconstruction of its 3D geometry. Each of the two recently proposed methods for face identification from line drawings involves two combinatorial problems. When dealing with complex objects having more faces, the combinatorial explosio... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Brightness perception, dynamic range and noise: a unifying model for adaptive image sensors

    Publication Year: 2004, Page(s):II-189 - II-196 Vol.2
    Cited by:  Papers (9)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1834 KB) | HTML iconHTML

    Many computer vision applications have to cope with large dynamic range and changing illumination conditions in the environment. Any attempt to deal with these conditions at the algorithmic level alone are inherently difficult because of the following: (1) conventional image sensors cannot completely capture wide dynamic range radiances without saturation or underexposure; (2) the quantization pro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.