Guillermo Cisneros

Audience Measurement Modeling for Convergent Broadcasting and IPTV Networks

Audience research is a vital part of TV and radio broadcasting, as well as of the more recent forms of media content delivery, such as the Internet, IPTV, mobile phones, Personal Video Recorders (PVRs) and portable media viewers. The uses of audience research range from self-promotion to refining service offerings and setting advertising rates. Without reliable audience data, many businesses will be reluctant to participate in the new platforms. This paper describes an end-to-end system for convergent audience measurement focused on IPTV but covering also terrestrial, cable, satellite and mobile broadcasting. We created the audience measurement system from the elaboration of a logical architectural model and a common data model which can be applied to any media scenario. We implemented this logical and data model in stationary and mobile media receivers (in the paper the particular case of IPTV is extensively explained). In addition user consumption is modeled and metrics are provided for user media consumption profiling and impact quantification in IPTV environments.

Iterative desensitisation of image restoration filters under wrong PSF and noise estimates

The restoration achieved on the basis of a Wiener scheme is an optimum since the restoration filter is the outcome of a minimisation process. Moreover, the Wiener restoration approach requires the estimation of some parameters related to the original image and the noise, as well as knowledge about the PSF function. However, in a real restoration problem, we may not possess accurate values of these parameters, making results relatively far from the desired optimum. Indeed, a desensitisation process is required to decrease this dependency on the parameter errors of the restoration filter. In this paper, we present an iterative method to reduce the sensitivity of a general restoration scheme (but specified to the Wiener filter) with regards to wrong estimates of the said parameters. Within the Fourier transform domain, a sensitivity analysis is tackled in depth with the purpose of defining a number of iterations for each frequency element, which leads to the aimed desensitisation regardless of the errors on estimates. Experimental computations using meaningful values of parameters are addressed. The proposed technique effectively achieves better results than those obtained when using the same wrong estimates in the Wiener approach, as well as verified on an SAR restoration.

User interest modeling for social TV-recommender systems based on audiovisual consumption

This paper presents an innovative consumption-modeling system applied to digital TV, able to infer the user interest regarding the TV contents. The user interest inferred is used to feed content recommender systems, and it has been tested in real scenarios involving real users. The modeling system uses as input the TV consumption data and performs an algorithm based on a Hidden Markov Model and Bayesian inference techniques to infer the said user interest. Real data have been picked up in real time to feed our modeling, and the final results have been checked comparing with user’s tastes, which have been expressed through a set of questionnaires and the whole system has been tested in a TV broadcasting scenario with real users. Conclusions show that our system improves the reliability from classic user interest modeling systems (they are mainly based on explicit opinion based methods, which can be intrusive for general users and could show certain deceptive results, as it is described in this paper).

Integration of Gabor functions with co-occurrence matrices: application to woody crop location in remote sensing

This paper proposes the combined use of Multichannel Filtering using Gabor Functions and Gray Level Co-occurrence Matrices (MCF/GF-GLCM) for descriptor extraction. This technique has been applied to location of woody crops in remotely sensed images. The use of GLCM introduces important improvements with respect to amplitude envelopment (modulus function). The test bed consisted of high spatial resolution images such as aerial photography and those images scanned by the Daedalus-1268 multispectral scanner on board of an airborne platform, in which textons (trees) are shown as isolated elements.

Remote sensing segmentation through a filter bank based on Gabor functions

One of the most critical activities of remote sensing consists of the identification of different coverages (types of crops) on land surface. This task is complicated when the entire plot is not covered by the crop (e.g. almond and olive fields, vineyards, etc.). The segregation of these crops is accomplished by using a multi-channel (Gabor functions) filtering approach in remote sensing imagery, in this case applied to an aerial photograph.

Developing Multimedia Applications: System Modeling and Implementation

This paper presents a model to develop novel user applications (or services) dealing with Multimedia Information retrieval through Telecom Networks, using a core application subsystem. Once the core system is available at the Information Provider sites, different applications can be easily developed on top of the services provided by the system. A building blocks architecture is proposed to provide access to storage systems, independently of their internal per site structure. The work carried out under the RAMA Project is also described.

From multimedia stream models to GUI generation

This paper is centered on the description of a model that generalizes multimedia data flows handling including complete behavior and interaction mechanisms, hence allowing full integration of GUIs generation -- GUI components are upgraded to interactive media items -- into the same unified model. It aims to reinforce portability, reusability, and quick development of multimedia applications. A picture of previous and of current state-of-art in multimedia application development clearly shows the need for standard abstractions in this field. Current work in this direction leads to a discussion on generic application structure (objects, semantics, etc.) and on different approaches to reach platform independence and efficient object sharing (formal representation languages, interpreted programming languages, distributed environments, etc.). At this point, we present a basic model based on several stream-based models and implementations on multimedia data flows, and built on the basis of the source-stream-sink paradigm. It follows with a detailed explanation of the unified (common to all media) abstract basic stream from which all monomedia flows (including GUI elements) are derived: stream setting- up (source-sink adaptation, and negotiation), flow control procedure, stream sensibility, behavior pattern, etc. The model presentation ends up with the introduction of the multimedia stream that performs synchronization and inter- stream communication tasks, and channels all sensibility, from/towards its managed streams, and allows for the design of an application generator. Then it deeps into the definition of the abstract class hierarchy that guides the model implementation. Finally, several implementation issues are addressed and some practical achievements are described.

Integration of Multisensorial Stimuli and Multimodal Interaction in a Hybrid 3DTV System

This article proposes the integration of multisensorial stimuli and multimodal interaction components into a sports multimedia asset under two dimensions: immersion and interaction. The first dimension comprises a binaural audio system and a set of sensory effects synchronized with the audiovisual content, whereas the second explores interaction through the insertion of interactive 3D objects into the main screen and on-demand presentation of additional information in a second touchscreen. We present an end-to-end solution integrating these components into a hybrid (internet-broadcast) television system using current 3DTV standards. Results from an experimental study analyzing the perceived quality of these stimuli and their influence on the Quality of Experience are presented.

An MLP neural net with L1 and L2 regularizers for real conditions of deblurring

Real conditions of deblurring involve a spatially nonlinear process since the borders are truncated, causing significant artifacts in the restored results. Typically, it is assumed to have boundary conditions to reduce ringing; in contrast, this paper proposes a restoration method which simply deals with null borders. We minimize a deterministic regularized function in a Multilayer Perceptron (MLP) with no training and follow a back-propagation algorithm with the L1 and L2 norm-based regularizers. As a result, the truncated borders are regenerated while adapting the center of the image to the optimum linear solution. We report experimental results showing the good performance of our approach in a real model without borders. Even if using boundary conditions, the quality of restoration is comparable to other recent researches.

Audience measurement technologies for user centric media

How to know which is the consumption and generation of content in a new media world where the user can be a consumer or a creator/distributor, compared to the traditional media? How interesting is it for the final users? Which new viable business models can be found in this world or which is the possible evolution of the market and the technology? Possibly, it is rather impossible to answer these questions and possibly never will be answered properly but the development of new audience measurement reference models for the new media world and its adaptation to the user centric media using combined metrics, are the most promising ways of achieving these goals. This paper describes an end-to-end system proposal based on the said reference model for the new media world applied to user centric media to provide an answer to some of the above questions and to obtain the necessary resulting figures.