Showing papers in "Image and Vision Computing in 2009"

TL;DR: This paper empirically evaluates facial representation based on statistical local features, Local Binary Patterns, for person-independent facial expression recognition, and observes that LBP features perform stably and robustly over a useful range of low resolutions of face images, and yield promising performance in compressed low-resolution video sequences captured in real-world environments.

TL;DR: A novel algorithm is introduced which allows us to approximately, or suboptimally, compute edit distance in a substantially faster way and is emprically verified that the accuracy of the suboptimal distance remains sufficiently accurate for various pattern recognition applications.

TL;DR: To enhance image detection rate and simplify computation of image retrieval, sequential forward selection is adopted for feature selection and three image databases with different properties are used to carry out feature selection.

TL;DR: This paper explores an approach for automatically recognizing acute pain without the need for human observers in adult patients with rotator cuff injuries and explored two questions pertinent to the construction, design and development of automatic pain detection systems.

TL;DR: This paper proposes a perceptual quality evaluation method for image fusion which is based on human visual system (HVS) models and finds the algorithm provides better predictions, which are more closely matched to human perceptual evaluations, than the existing algorithms.

TL;DR: This paper proposes a novel method capable of dividing an investigated image into various partitions with homogenous noise levels and introduces a segmentation method detecting changes in noise level using the additive white Gaussian noise.

TL;DR: This paper reviews the existing literature on automatic analysis of small group conversations using nonverbal communication, and aims at bridging the current fragmentation of the work in this domain, currently split among half a dozen technical communities.

TL;DR: A local bundle adjustment is introduced allowing 3D points and camera poses to be refined simultaneously through the sequence, which significantly reduces computational complexity when compared with global bundle adjustment.

TL;DR: The automated system was successfully able to differentiate faked from real pain, and the most discriminative facial actions in the automated system were consistent with findings using human expert FACS codes.

TL;DR: It is demonstrated that high precision can be achieved by combining multiple sources of information, both visual and textual, by automatic generation of time stamped character annotation by aligning subtitles and transcripts.

TL;DR: A novel defect detection scheme based on morphological filters is proposed to tackle the problem of automated defect detection for woven fabrics using a pre-trained Gabor wavelet network to remove the fabric background and isolate the defects.

TL;DR: The purpose of the proposed system is to assist in a biometric authentication framework, by adding liveness awareness in a non-intrusive manner, by analyzing the trajectories of certain parts of a live face to discriminate it against a spoofed one.

TL;DR: This work defines a well-localized edge landmark and presents an efficient algorithm for selecting such landmarks and describes how to initialize new landmarks, observe mapped landmarks in subsequent images, and address the data association challenges of edges.

TL;DR: A fully automatic processing combination of Active-Appearance-Model-based facial expression, vision-based eye-activity estimation, acoustic features, linguistic analysis, non-linguistic vocalisations, and temporal context information in an early feature fusion process is introduced.

TL;DR: This paper defines a new error measure which quantifies the performance of an image segmentation algorithm for identifying multiple objects in an image based on object-by-object comparisons of a segmented image and a ground-truth (reference) image.

TL;DR: The proposed Graph Transformation Matching method, relying on finding a consensus nearest-neighbour graph emerging from candidate matches, is successfully applied in the context of constructing mosaics of retinal images, where feature points are extracted from properly segmented binary images.

TL;DR: In this article, the security of a chaos-based image encryption scheme was studied and it was shown that there exist a number of invalid keys and weak keys, and some keys are partially equivalent for encryption/decryption.

TL;DR: A new idea of curvelet based local binary pattern is proposed as textural features to distinguish ulcer regions from normal regions, which makes full use of Curvelet transformation and local binary patterns.

TL;DR: A novel approach for combining a set of registered images into a composite mosaic with no visible seams and minimal texture distortion is presented, which allows the efficient creation of large mosaics, without user intervention.

TL;DR: The results of experiments confirm the robustness of proposed method against severe imaging conditions and the ability to characterize the color information in plate using the MNS (multimodal neighborhood signature) method.

TL;DR: This work proposes a novel method for enhancing watershed segmentation by utilizing prior shape and appearance knowledge, which iteratively aligns a shape histogram with the result of an improved k-means clustering algorithm of the watershed segments.

TL;DR: A system that can automatically track multiple hockey players and simultaneously recognize their actions given a single broadcast video sequence, where detection is complicated by a panning, tilting, and zooming camera is presented.

TL;DR: The security of the proposed image encryption scheme is studied and the following problems are found: a differential chosen-plaintext attack can break the scheme with only three chosen plain-images; there is a number of weak keys and some equivalent keys for encryption; the scheme is not sensitive to the changes of plain- images.

TL;DR: This study shows that, human actions can be simply represented by pose without dealing with the complex representation of dynamics, and proves that with a simple and compact representation, this system can achieve robust recognition of human actions, compared to complex representations.

TL;DR: A novel method for employing image-based rendering to extend the range of applicability of human motion and gait recognition systems by automatically constructing the proper view (image) that matches the training view, from a combination of arbitrary views taken from several cameras.

TL;DR: This work shows how histogram-based image descriptors can be combined with a boosting classifier to provide a state of the art object detector and introduces a weak learner for multi-valued histogram features and shows how to overcome problems of limited training sets.

TL;DR: This paper uses a rule-based method to propagate the vessel labels through the vascular graph and embeds this task as double-layered constrained search problem steered by a heuristical AC-3 algorithm to overcome the NP-hard computational complexity.

TL;DR: This paper presents a face recognition algorithm that addresses two major challenges: when an individual intentionally alters the appearance and features using disguises, and when limited gallery images are available for recognition.

TL;DR: A new combination of techniques is presented to create pleasing and physically consistent image mosaics despite the presence of moving objects in the scene using heuristic seam selection in the intensity and gradient-domains.

TL;DR: This work presents an interactive segmentation algorithm that can segment an object of interest from its background with minimum guidance from the user, who just has to select a single seed pixel inside the object ofinterest.