Showing papers in "Pattern Recognition Letters in 2009"

TL;DR: The Cambridge-driving Labeled Video Database (CamVid) is presented as the first collection of videos with object class semantic labels, complete with metadata, and the relevance of the database is evaluated by measuring the performance of an algorithm from each of three distinct domains: multi-class object recognition, pedestrian detection, and label propagation.

TL;DR: This work analyzes experimentally the behaviour of 18 different performance metrics in several scenarios, identifying clusters and relationships between measures and makes a comprehensive analysis of the relationships between metrics, and a taxonomy and arrangement of them according to the previous traits.

TL;DR: The proposed method depends on a single parameter only and can be used for characterising binary patterns with emphasis on connections between their parts as measured at varying analysis scales.

TL;DR: This paper proposes a novel feature extraction algorithm, namely binary orientation co-occurrence vector (BOCV), to represent multiple orientations for a local region, which can better describe the local orientation features and it is more robust to image rotation.

TL;DR: This paper investigates feature subset selection for dimensionality reduction in machine learning and finds that GRASP and Tabu Search obtain significantly better results than the other methods.

TL;DR: The experimental results show that the proposed TSMO method is far more efficient with an accuracy equivalent to Otsu's method, and has the advantage of having a small variance in runtimes for different test images.

TL;DR: Experimental results show the superiority of FDEI representation over binary silhouettes and some other algorithms when occlusion or body portion lost appears in the gait sequences.

TL;DR: The testing results on FIFA World Cup 2006 videos demonstrate that the method can reach high detection and labeling precision, and reliably tracking in cases of scenes such as player occlusion, moderate camera motion and pose variation.

TL;DR: This work proposes an extended Neighbor embedding based super-resolution through edge detection and Feature Selection (henceforth NeedFS), which is robust even with a very limited training set and thus is promising for real applications.

TL;DR: This paper constructs color invariant filter sets from the original MR8 filterbank and demonstrates the material-specific filterbank models to be preferred over models with fixed filterbanks.

TL;DR: The proposed hybrid clustering technique called rough-DBSCAN has a time complexity of O(n) only and is analyzed using rough set theory and can find a similar clustering as found by the DBSCAN, but is consistently faster than DBS CAN.

TL;DR: The LBP-TOP features are extended to multi-resolution spatiotemporal space and used for describing facial expressions and AdaBoost is utilized to learn the principal appearance and motion and for selecting the most important expression-related features.

TL;DR: This paper presents a new formulation of the ternary ECOC distance and the error-correcting capabilities in the terrifying ECOC framework, and stresses on how to design coding matrices preventing codification ambiguity and proposes a new sparse random coding matrix with ternARY distance maximization.

TL;DR: This paper proposes a new algorithm, CamShift guided particle filter (CAMSGPF), which can track objects robustly in various environments, and is much faster than the existing methods.

TL;DR: The new signature of integral of unsigned curvatures is presented as well as faster algorithms for matching open 2D curves and examples from diverse application set are presented to show that the algorithm can work across several domains.

TL;DR: This paper designs the dynamic haar-like features to represent the temporal variations of facial appearance and further encode the dynamic features into the binary pattern features, which are useful to construct weak classifiers for boosting learning.

TL;DR: An attempt to deal with the problem of temporal overlaps at the level of models using the SVM-based AED system is reported in the paper, and the proposed two-step system noticeably outperforms the baseline system for both an artificially generated database and a real seminar recording database.

TL;DR: A complete and efficient video in-painting system that applies different strategies to handle static and dynamic portions of the hole and evaluates the performance of the system based on a set of indoor surveillance sequences with different types of occlusions.

TL;DR: A novel approach to UAV's automatic landing on the ship's deck is proposed, the design of the cooperative object is presented, and basic research on UAV autonomous landing on a ship is begun by using computer vision and affine moment invariants.

TL;DR: This work addresses the problem of face recognition from image sets, where subject-specific subspaces instead of image vectors are compared, and shows that kernel Grassmannian distances in feature space can be implicitly computed from the input data.

TL;DR: It is shown that when the filter response vectors are quantized for histogram based joint density estimation, thresholding is clearly faster than using learned codebooks and, being robust to gray-level changes, it yields better recognition rate in most cases.

TL;DR: A standard one-class SVM is reformulated and a least squares version of the method is derived, which is called LS (least squares) one- class SVM, which demonstrates the performance of the LS one- Class SVM on relevance ranking with positive examples, and also presents the comparison with traditional methods.

TL;DR: An improved EM method is presented to model regional color distribution of the target image by Gaussian Mixture Model, then appropriate reference colors are automatically selected from the given source images to color each target region.

TL;DR: This paper attempts to define and segment subunits using computer vision techniques, which also can be basically explained by sign language linguistics and correlates highly with the definition of syllables in sign language while sharing characteristics of syllable in spoken languages.

TL;DR: A symbol shape description to deal with the changes in appearance that these types of symbols suffer, and the Blurred Shape Model descriptor (BSM), where new features encode the probability of appearance of each pixel that outlines the symbols shape.

TL;DR: A new feature scaling method, called class-dependent-feature-weighting (CDFW) using naive Bayes (NB) classifier, which combines CDFW and recursive feature elimination (RFE) and results showed that CDFW-NB-RFE outperformed other popular feature ranking schemes used on text datasets.

TL;DR: A multiscale texture classifier that exploits the Gabor-like properties of the dual-tree complex wavelet transform, shift invariance and six directional sub bands at each scale, and uses a feature vector comprising of a variance and an entropy at different scales of each of the directional subbands.

TL;DR: Experimental results on extended Yale face database B and CMU PIE face databases demonstrate that ESLLE obtains better performance on face recognition compared with other famous methods such as principal component analysis (PCA), locally linear embedding (LLE), as well as SLLE.

TL;DR: A stroke filter which can detect strokes of texts for robust text localization and remove text candidates which have strong edges but are not text is proposed.

TL;DR: This paper shows the advantages of using smooth derivative filters instead of pixel differences in the performance of a well known local image descriptor, and increases the distinctiveness of local image descriptors for image region matching and object recognition.