Showing papers in "Pattern Recognition in 2015"

TL;DR: The independence assumptions in cross validation are introduced, and the circumstances that satisfy the assumptions are also addressed, which are used to derive the sampling distributions of the point estimators for k-fold and leave-one-out cross validation.

TL;DR: A scalable distance driven feature learning framework based on the deep neural network for person re-identification that achieves very promising results and outperforms other state-of-the-art approaches.

TL;DR: A survey of Hough Transform and its variants, their limitations and the modifications made to overcome them, the implementation issues in software and hardware, and applications in various fields is done.

TL;DR: A novel ensemble method is proposed, which firstly converts an imbalanced data set into multiple balanced ones and then builds a number of classifiers on these multiple data with a specific classification algorithm, which usually performs better than the conventional methods on im balanced data.

TL;DR: This paper provides an efficient EMR-SLRA optimization procedure to obtain the output feature embedding and experiments on the pattern recognition applications confirm the effectiveness of the EMR -SLRA algorithm compare with some other multiview feature dimensionality reduction approaches.

TL;DR: Experimental analysis on synthetic and real-world data demonstrates that the proposed regularized self-representation (RSR) model can effectively identify the representative features, outperforming many state-of-the-art unsupervised feature selection methods in terms of clustering accuracy, redundancy reduction and classification accuracy.

TL;DR: This paper proposes a novel computational visual attention model, namely saliency structure model, for content-based image retrieval, and introduces a novel visual cue, namely color volume, with edge information together, to detect saliency regions.

TL;DR: This work proposes to exploit a pre-trained large convolutional neural network to generate deep features for CRF learning and construct spatially related co-occurrence pairwise potentials and incorporate them into the energy function to fully exploit context information in inference.

TL;DR: An overview of state-of-the-art methods in activity recognition using semantic features, including a semantic space including the most popular semantic features of an action namely the human body, attributes, related objects, and scene context, is presented.

TL;DR: The qualitative and quantitative comparisons on two publicly available databases demonstrate that the proposed regularized Gaussian fields criterion for non-rigid registration of visible and infrared face images significantly outperforms the state-of-the-art method with an affine model.

TL;DR: A novel algorithm that combines sparse and collaborative representation is proposed for target detection in hyperspectral imagery that outperforms the existing target detection algorithms, such as adaptive coherence estimator and pure sparse representation-based detector.

TL;DR: The intent is to represent skeletal motion in a geometric and efficient way, leading to an accurate action-recognition system, which represents skeletal sequence as point on the Grassmann manifold.

TL;DR: This paper proposes the use of One-Class Support Vector Machine (OC-SVM) based on writer-independent parameters, which takes into consideration only genuine signatures and when forgery signatures are lack as counterexamples for designing the HSVS.

TL;DR: Experimental results show that the proposed meta-learning framework greatly improves classification accuracy when compared against current state-of-the-art dynamic ensemble selection techniques.

TL;DR: This paper formulate the video summarization task with a novel minimum sparse reconstruction (MSR) problem, where the original video sequence can be best reconstructed with as few selected keyframes as possible.

TL;DR: A Hidden Markov Model based approach for detecting abnormalities in daily activities, a process of identifying irregularity in routine behaviours from statistical histories and an exponential smoothing technique to predict future changes in various vital signs are described.

TL;DR: Experimental results on FERET, AR, Yale face databases and the PolyU finger-knuckle-print database demonstrate that CRP works well in feature extraction and leads to a good recognition performance.

TL;DR: Unsupervised and multivariate filter-based feature selection methods are proposed by analyzing the relevance and redundancy of features by using ant colony optimization and a novel heuristic information measure is proposed to enhance the accuracy of the methods.

TL;DR: Two robust statistical techniques for outlier detection and robust saliency features, such as surface normal and curvature, estimation in laser scanning 3D point cloud data, based on a robust z-score and a Mahalanobis type robust distance are proposed.

TL;DR: The experimental results show that the proposed neutrosophic c -means algorithm can be considered as a promising tool for data clustering and image processing.

TL;DR: The results on the eight real world datasets indicate that IWFS not only efficiently reduces the dimensionality of feature space, but also offers the highest average accuracy for all the three classification algorithms.

TL;DR: Experiments on the publicly available LivDet 2011 database, comprising datasets collected from various sensors, prove the proposed method to outperform the state-of-the-art liveness detection techniques.

TL;DR: A new hierarchical decomposition method for imbalanced data sets which is different from previously proposed solutions to the class imbalance problem and does not require any data pre-processing step as many other solutions need.

TL;DR: A new unsupervised feature selection criterion developed from the viewpoint of subspace learning, which is treated as a matrix factorization problem and which provides a sound foundation for embedding kernel tricks into feature selection problems.

TL;DR: The results demonstrate that not only idle-state decoding can be significantly improved by exploiting the complementary information of multi-modal recordings, but also it is possible to minimize the delay of the system, caused by the slow inherent hemodynamic response of the NIRS signal.

TL;DR: An efficient segmentation-free word spotting method, applied in the context of historical document collections, that follows the query-by-example paradigm that outperforms the recent state-of-the-art keyword spotting approaches.

TL;DR: A robust global and local mixture distance (GLMD) based non-rigid point set registration method which consists of an alternating two-step process: correspondence estimation and transformation updating and which outperforms state-of-the-art methods in this area.

TL;DR: A quadratic time approximation of graph edit distance based on Hausdorff matching is proposed and shows a promising potential in terms of flexibility, efficiency, and accuracy.

TL;DR: A general bag dissimilarities framework for multiple instance learning is explored and several alternatives to define a dissimilarity between bags are shown and discussed, which definitions are more suitable for particular MIL problems.

TL;DR: A coding-based algorithm for salient object detection that outperforms 22 state-of-the-art methods in terms of three popular evaluation measures, i.e., the Precision and Recall curve, Area Under ROC Curve and F-measure value.