Showing papers in "Computer Vision and Image Understanding in 2005"

TL;DR: A detailed description of the pupil-corneal reflection technique is presented due to its claimed usability advantages, and it is shown that this method is still not quite appropriate for general interactive applications.

TL;DR: This paper provides an up-to-date review of research efforts in face recognition techniques based on two-dimensional images in the visual and infrared (IR) spectra.

TL;DR: Results show that the triangulation algorithm outperforms standard linear and bias-corrected quasi-linear algorithms, and that bundle adjustment using the orthonormal representation yields results similar to the standard maximum likelihood trifocal tensor algorithm, while being usable for any number of views.

TL;DR: This work confirms not only that ears do indeed appear to have potential as a biometric, but also that the new approach is well suited to their description, being robust especially in the presence of noise, and having the advantages that the ear does not need to be explicitly extracted from the background.

TL;DR: An integrated eye tracker that combines the latest technologies in appearance-based object recognition and tracking with active IR illumination that can robustly track eyes under variable and realistic lighting conditions and under various face orientations is presented.

TL;DR: To estimate where the user looks allowing ample head movement, an invariant value (cross-ratio) of a projective space is utilized and a robust feature detection using an ellipse-specific active contour is suggested in order to find features exactly.

TL;DR: This paper presents an algorithm for modeling, tracking, and recognizing human faces in video sequences within one integrated framework that emphasizes an algorithmic architecture that tightly couples these two components within a single framework.

TL;DR: The proposed method for the selection of salient regions likely to contain objects, based on bottom-up visual attention, can enable one-shot learning of multiple objects from complex scenes, and can strongly improve learning and recognition performance in the presence of large amounts of clutter.

TL;DR: A probabilistic model of image generation is formulated and optimal inference algorithms for finding objects and object features within this framework are derived, which works in real time and is robust to changes in lighting, illumination, and differences in facial structure.

TL;DR: This work presents a novel approach for personal identification and identity verification which utilizes 3D finger surface features as a biometric identifier using the curvature based shape index to represent the fingers' surface.

TL;DR: An active contour tracker is presented which can be used for gaze-based interaction with off-the-shelf components and a lower bound on the number of calibration points needed for gaze determination on planar objects is presented.

TL;DR: Experimental results reinforce prior research that employed a smaller data set, presenting a convincing argument that, even across a broad experimental spectrum, the behaviors enumerated above are valid and consistent.

TL;DR: This paper proposes a level set approach to address both aspects of motion analysis, which relies on the propagation of smooth interfaces to perform tracking while using an incremental estimation of the motion models.

TL;DR: The problems of achieving a dense reconstruction from a sequence of images and analyzing and removing specular highlights are addressed, and a taxonomy of specularities based on their photometric properties is presented as a guide for designing further separation techniques.

TL;DR: An in-depth analysis of the saliency-based model of visual attention by assessing the contribution of different cues to visual attention as modeled by different versions of the computer model is presented.

TL;DR: The two key contributions in this paper are to show that it is possible to estimate eye gaze with only one eye image and that consequently this achieves higher accuracy of eye gaze estimation.

TL;DR: A novel system to detect houses and street networks in IKONOS multispectral images with one meter panchromatic resolution with 4 m resolution in the spectral bands is introduced, indicating the usefulness of the system in detecting houses andStreet networks, hence generating automated maps.

TL;DR: This paper presents an approach for automatic extraction of dominant rectangular structures from a single view and shows how they facilitate the recovery of camera pose, planar structure and matching across widely separated views.

TL;DR: A novel model of attentive visual motion processing is presented that addresses both decomposition of the signal into constituent features as well as the re-combination, or binding, of those features into wholes.

TL;DR: The hardware implementation of a high complexity algorithm to estimate the optical flow from image sequences in real time is described, which has many applications in fields like object recognition, image segmentation, autonomous navigation, and security systems.

TL;DR: This work suggests an approach in which high-level cognitive processes are top-down directed and modulate stimulus signals such that vision is a constructive process in time.

TL;DR: A novel design for a non-contact eye detection and gaze tracking device that uses two cameras to maintain real-time tracking of a person's eye in the presence of head motion and three methods of estimating the user's point of gaze on a computer monitor are evaluated.

TL;DR: In this article, the problem of finding optimal point correspondences between images related by a homography is addressed, and the problem is reduced to the solution of a polynomial of degree eight in a single variable, which can be computed numerically.

TL;DR: A new concept called "tensor scale" is introduced--a local morphometric parameter yielding a unified representation of structure size, orientation, and anisotropy in anisotropic diffusive image filtering that encourages smoothing inside a homogeneous region and also along edges and elongated structures while discourages blurring across them.

TL;DR: Two new methods to label connected components based on iterative recursion are proposed: one directly labels an original binary image while the other labels the boundary voxel followed by one-pass labelling of non-boundary object voxels.

TL;DR: This paper investigates how disjoint pairs of bearing or range sensors can be best assigned to targets to minimize the expected error in the resulting estimation for target locations, and obtains significant approximation results for specific geometries.

TL;DR: In this article, the authors present a framework for analyzing, comparing, developing, and implementing surface correspondence algorithms, which can be used to analyse and compare existing algorithms and develop new algorithms using the framework's modular structure.

TL;DR: The aim of this paper is to propose a new parameter estimation method adapted to TMF, and to study the corresponding unsupervised image segmentation methods.

TL;DR: This paper presents a hardware-based embedded system for eye detection, implemented using simple logic gates, with no CPU and no addressable frame buffers, suitable for single-chip eye detection and eye-gaze tracking sensors, thus making an important step towards mass production, low cost systems.

TL;DR: The fundamental idea of the proposed LDI is based on an observation that the ratios of orientations and location parameters of left- and right-lane boundaries are equal to one as far as the optical axis of a camera mounted on a vehicle is coincident with the center of lane.