Showing papers in "Computer Vision and Image Understanding in 2007"

TL;DR: The incremental algorithm is compared experimentally to an earlier batch Bayesian algorithm, as well as to one based on maximum-likelihood, which have comparable classification performance on small training sets, but incremental learning is significantly faster, making real-time learning feasible.

TL;DR: This paper reviews the major approaches to multimodal human-computer interaction, giving an overview of the field from a computer vision perspective, and focuses on body, gesture, gaze, and affective interaction.

TL;DR: The characteristics of human motion analysis are discussed to highlight trends in the domain and to point out limitations of the current state of the art.

TL;DR: A literature review on the second research direction, which aims to capture the real 3D motion of the hand, which is a very challenging problem in the context of HCI.

TL;DR: An original method for grading the colours between different images or shots is proposed using an original and parameter free algorithm that is able to transform any N-dimensional probability density function into another one.

TL;DR: A new background-subtraction technique fusing contours from thermal and visible imagery for persistent object detection in urban settings is presented, evaluated quantitatively and compared with other low- and high-level fusion techniques using manually segmented data.

TL;DR: This research aims to realize the notion of desktop health monitoring and create truly collaborative interactions in which humans and machines are both observing and responding.

TL;DR: In this article, a graph cuts based active contours (GCBAC) approach is proposed for object segmentation, which uses graph cuts to iteratively deform the contour and its cost function is defined as the summation of edge weights on the cut.

TL;DR: A systematic experimental evaluation on a large video database with human actions demonstrates that local spatio-temporal image descriptors can be defined to carry important information of space-time events for subsequent recognition and that local velocity adaptation is an important mechanism in situations when the relative motion between the camera and the interesting events in the scene is unknown.

TL;DR: This work manifests the best algorithms suited for Gaussian and mean curvature estimation, and shows that different algorithms should be employed to compute the Gaussianand mean curvatures.

TL;DR: A generalisation of distance-based and lexicographical-based approaches is proposed, allowing the extension of morphological operators to colour images for any colour representation and for any metric distance to a reference colour.

TL;DR: A layered representation is first introduced and a generalized expectation-maximization (EM) algorithm is developed to separate infrared images into background (still) and foreground (moving) layers regardless of camera panning to facilitate pedestrian tracking.

TL;DR: A generative image model, which is called ''primal sketch,'' following Marr's insight and terminology is proposed, which combines two prominent classes of generative models, namely, sparse coding model and Markov random field model for representing geometric structures and stochastic textures, respectively.

TL;DR: A stereo system for the detection of pedestrians using far-infrared cameras that exploits three different detection approaches: warm area detection, edge-based detection, and disparity computation.

TL;DR: A new theory and an algorithm for image segmentation based on a strength of connectedness between every pair of image elements based on the notion of iterative relative fuzzy connectedness, IRFC, is presented.

TL;DR: This paper introduces a novel camera mouse driven by visual face tracking based on a 3D model which can be an alternative input option for people with hand and speech disability, as well as for futuristic vision-based games and interfaces.

TL;DR: This paper concludes that thermal Imagery provides relevant information for FER, and that the developed methodology can be taken as an efficient learning mechanism for different types of pattern recognition problems.

TL;DR: The proposed quaternion color texture representation is demonstrated by its use in an unsupervised segmentation algorithm that successfully divides an image into regions on basis of texture.

TL;DR: This work proposes an enhanced algorithm that wraps a particle filter around multiple SMD-based trackers, which play the role of many particles, i.e. that act as 'smart particles'.

TL;DR: A computer-based system that greatly augments the physician's vision during bronchoscopy, and the results indicate that not only is the variation in skill level between different physicians greatly reduced by the system over the standard procedure, but that biopsy effectiveness increases.

TL;DR: A novel color image histogram equalization approach is proposed that exploits the correlation between color components and it is enhanced by a multi-level smoothing technique borrowed from statistical language engineering in order to eliminate the gamut problem.

TL;DR: The RNDA relaxes Gaussian assumptions of Fisher discriminant analysis (FDA), and it can handle more general class distributions, and it improves the traditional nonparametric discriminantAnalysis (NDA) by alleviating its computational complexity.

TL;DR: A topographic modeling approach to recognize and analyze facial expression from single static images, developed based on a novel facial expression descriptor, Topographic Context (TC), and investigates the robustness of the extracted TC-based expression features.

TL;DR: This work proposes a selective metric-based clustering to extract textual information in real-world images based on physical light reflectance, thus enabling the processing of character segmentation into individual components to increase final recognition rates.

TL;DR: Ground truth experiments demonstrate the utility of the disparity voting techniques for multimodal registration by yielding qualitative and quantitative results that outperform approaches that do not consider occlusions.

TL;DR: A novel approach to recognizing driver activities using a multi-perspective multi-modal video-based system for robust and real-time tracking of important body parts and a Hidden Markov model-based classifier provides semantic-level analysis of human activity.

TL;DR: The proposed method is much faster than previous works, and can overcome tracking errors resulted from mixed data of two closely situated persons, and achieves a significant improvement compared with existing laser-based trackers.

TL;DR: This paper defines the conditions on the spectrum coefficients needed to reconstruct a colour image without loss of information through the inverse quaternionic Fourier transform process and defines spatial and frequential strategies to filter colour images.

TL;DR: The experimental results show that the proposed method for embedding a color or a grayscale image in a true color image is a secure steganographic method that provides high hiding capacity and good image quality.