Dounia Awad, M. Mancas, Nicolas Riche, V. Courboulay, A. Revel
{"title":"A CBIR-based evaluation framework for visual attention models","authors":"Dounia Awad, M. Mancas, Nicolas Riche, V. Courboulay, A. Revel","doi":"10.1109/EUSIPCO.2015.7362639","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362639","url":null,"abstract":"The computational models of visual attention, originally proposed as cognitive models of human attention, nowadays are being used as front-ends to numerous vision systems like automatic object recognition. These systems are generally evaluated against eye tracking data or manually segmented salient objects in images. We previously showed that this comparison can lead to different rankings depending on which of the two ground truths is used. These findings suggest that the saliency models ranking might be different for each application and the use of eye-tracking rankings to choose a model for a given application is not optimal. Therefore, in this paper, we propose a new saliency evaluation framework optimized for object recognition. This paper aims to answer the question: 1) Is the application-driven saliency models rankings consistent with classical ground truth like eye-tracking? 2) If not, which saliency models one should use for the precise CBIR applications?.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132746869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Santiago Segarra, A. Marques, G. Leus, Alejandro Ribeiro
{"title":"Interpolation of graph signals using shift-invariant graph filters","authors":"Santiago Segarra, A. Marques, G. Leus, Alejandro Ribeiro","doi":"10.1109/EUSIPCO.2015.7362375","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362375","url":null,"abstract":"New schemes to recover signals defined in the nodes of a graph are proposed. Our focus is on reconstructing bandlimited graph signals, which are signals that admit a sparse representation in a frequency domain related to the structure of the graph. The schemes are designed within the framework of linear shift-invariant graph filters and consider that the seeding signals are injected only at a subset of interpolating nodes. After several sequential applications of the graph-shift operator - which computes linear combinations of the information available at neighboring nodes - the seeding signals are diffused across the graph and the original bandlimited signal is eventually recovered. Conditions under which the recovery is feasible are given, and the corresponding schemes to recover the signal are proposed. Connections with the classical interpolation in the time domain are also discussed.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124225483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Low complexity single microphone tonal noise reduction in vehicular traffic environments","authors":"N. Chatlani, C. Beaugeant, P. Kroon","doi":"10.1109/EUSIPCO.2015.7362401","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362401","url":null,"abstract":"A low complexity single microphone Tonal Noise Reduction (TNR) technique is presented for speech enhancement. This method is particularly effective in noisy environments which contain tonal noise sources, such as vehicular horns and alarms. TNR was designed to have low complexity and low memory requirements for use with battery operated communication devices. TNR detects the probability of the presence of these tonal noises which contaminate the desired speech signals. These noises are then attenuated using the proposed system for noise suppression. This is particularly effective for noise sources with a harmonic spectral structure. The proposed TNR system is able to maintain a balance between the level of noise reduction and speech distortion. Listening tests were performed to confirm the results. TNR can be used together with a general noise reduction system as a postprocessing stage by reducing the residual noise components.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114612473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Paris V. Giampouras, A. Rontogiannis, K. Themelis, K. Koutroumbas
{"title":"Online Bayesian low-rank subspace learning from partial observations","authors":"Paris V. Giampouras, A. Rontogiannis, K. Themelis, K. Koutroumbas","doi":"10.1109/EUSIPCO.2015.7362840","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362840","url":null,"abstract":"Learning the underlying low-dimensional subspace from streaming incomplete high-dimensional observations data has attracted considerable attention lately. In this paper, we present a new computationally efficient Bayesian scheme for online low-rank subspace learning and matrix completion. The proposed scheme builds upon a properly defined hierarchical Bayesian model that explicitly imposes low rank to the latent subspace by assigning sparsity promoting Student-t priors to the columns of the subspace matrix. The new algorithm is fully automated and as corroborated by numerical simulations, provides higher estimation accuracy and a better estimate of the true subspace rank compared to state of the art methods.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114798932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Ultrasound compressive deconvolution with ℓP-Norm prior","authors":"Zhouye Chen, Ningning Zhao, A. Basarab, D. Kouamé","doi":"10.1109/EUSIPCO.2015.7362893","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362893","url":null,"abstract":"It has been recently shown that compressive sampling is an interesting perspective for fast ultrasound imaging. This paper addresses the problem of compressive deconvolution for ultrasound imaging systems using an assumption of generalized Gaussian distributed tissue reflectivity function. The benefit of compressive deconvolution is the joint volume reduction of the acquired data and the image resolution improvement. The main contribution of this work is to apply the framework of compressive deconvolution on ultrasound imaging and to propose a novel ℓp-norm (1 ≤ p ≤ 2) algorithm based on Alternating Direction Method of Multipliers. The performance of the proposed algorithm is tested on simulated data and compared with those obtained by a more intuitive sequential compressive deconvolution method.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114516653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"More efficient sparsity-inducing algorithms using inexact gradient","authors":"A. Rakotomamonjy, Sokol Koço, L. Ralaivola","doi":"10.1109/EUSIPCO.2015.7362475","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362475","url":null,"abstract":"In this paper, we tackle the problem of adapting a set of classic sparsity-inducing methods to cases when the gradient of the objective function is either difficult or very expensive to compute. Our contributions are two-fold: first, we propose methodologies for computing fair estimations of inexact gradients, second we propose novel stopping criteria for computing these gradients. For each contribution we provide theoretical backgrounds and justifications. In the experimental part, we study the impact of the proposed methods for two well-known algorithms, Frank-Wolfe and Orthogonal Matching Pursuit. Results on toy datasets show that inexact gradients can be as useful as exact ones provided the appropriate stopping criterion is used.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"217 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114982792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Dynamic speech emotion recognition with state-space models","authors":"K. Markov, T. Matsui, F. Septier, G. Peters","doi":"10.1109/EUSIPCO.2015.7362750","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362750","url":null,"abstract":"Automatic emotion recognition from speech has been focused mainly on identifying categorical or static affect states, but the spectrum of human emotion is continuous and time-varying. In this paper, we present a recognition system for dynamic speech emotion based on state-space models (SSMs). The prediction of the unknown emotion trajectory in the affect space spanned by Arousal, Valence, and Dominance (A-V-D) descriptors is cast as a time series filtering task. The state space models we investigated include a standard linear model (Kalman filter) as well as novel non-linear, non-parametric Gaussian Processes (GP) based SSM. We use the AVEC 2014 database for evaluation, which provides ground truth A-V-D labels which allows state and measurement functions to be learned separately simplifying the model training. For the filtering with GP SSM, we used two approximation methods: a recently proposed analytic method and Particle filter. All models were evaluated in terms of average Pearson correlation R and root mean square error (RMSE). The results show that using the same feature vectors, the GP SSMs achieve twice higher correlation and twice smaller RMSE than a Kalman filter.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116873478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ahmed Selloum, S. Hosseini, T. Contini, Y. Deville
{"title":"Semi-blind separation of galaxy spectra from a mixture obtained by slitless spectroscopy","authors":"Ahmed Selloum, S. Hosseini, T. Contini, Y. Deville","doi":"10.1109/EUSIPCO.2015.7362662","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362662","url":null,"abstract":"We investigate the problem of separating galaxy spectra from their mixtures resulting from the slitless spectroscopy used in the future Euclid space mission. This can be formulated as a source separation problem where the structure of the mixture is specific and depends on a low number of parameters. We first develop a mathematical model to describe the observations generated by the near-infrared spectrograph of Euclid, then propose non-blind, semi-blind and regularized semi-blind methods to separate the spectra. The first simulation results are encouraging: even for a signal to noise ratio of 5 dB, our regularized semi-blind method succeeds in separating the considered two spectra and provides a satisfactory estimate of the emission line positions and amplitudes.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"05 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121987565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fernando G. Almeida Neto, V. Nascimento, Amanda Souza de Paula
{"title":"Distributed multichannel adaptive filtering","authors":"Fernando G. Almeida Neto, V. Nascimento, Amanda Souza de Paula","doi":"10.1109/EUSIPCO.2015.7362569","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362569","url":null,"abstract":"A new distributed multichannel technique is proposed for networks in which a different set of parameters is estimated by each node. The technique is proposed for non fully-connect topologies, so that nodes must store data and re-transmit information to other network elements. To reduce the amount of terms stored by each node, pre-computation of the data required by other elements is performed before the data sharing. The proposed method is adequate for implementation in networks with a large number of nodes, for which straightforward implementations would be prohibitive in terms of cost and memory.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125996739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A novel deterministic method for large-scale blind source separation","authors":"Martijn Boussé, Otto Debals, L. D. Lathauwer","doi":"10.1109/EUSIPCO.2015.7362712","DOIUrl":"https://doi.org/10.1109/EUSIPCO.2015.7362712","url":null,"abstract":"A novel deterministic method for blind source separation is presented. In contrast to common methods such as independent component analysis, only mild assumptions are imposed on the sources. On the contrary, the method exploits a hypothesized (approximate) intrinsic low-rank structure of the mixing vectors. This is a very natural assumption for problems with many sensors. As such, the blind source separation problem can be reformulated as the computation of a tensor decomposition by applying a low-rank approximation to the tensorized mixing vectors. This allows the introduction of blind source separation in certain big data applications, where other methods fall short.","PeriodicalId":401040,"journal":{"name":"2015 23rd European Signal Processing Conference (EUSIPCO)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124814282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}