Content Description Methods Research Articles

Retrieval of Remote Sensing Images with Pattern Spectra Descriptors

The rapidly increasing volume of visual Earth Observation data calls for effective content based image retrieval solutions, specifically tailored for their high spatial resolution and heterogeneous content. In this paper, we address this issue with a novel local implementation of the well-known morphological descriptors called pattern spectra. They are computationally efficient histogram-like structures describing the global distribution of arbitrarily defined attributes of connected image components. Besides employing pattern spectra for the first time in this context, our main contribution lies in their dense calculation, at a local scale, thus enabling their combination with sophisticated visual vocabulary strategies. The Merced Landuse/Landcover dataset has been used for comparing the proposed strategy against alternative global and local content description methods, where the introduced approach is shown to yield promising performances.

The Design and Evaluation of an InformationSharing System for Human Networks

With fast-growing consumer demands and rapidly-developing mobile technologies, portable mobile devices are becoming a necessity of our daily lives. However, existing mobile devices rely on the wireless infrastructure to access Internet services provided by central application providers. This architecture is inefficient in many situations and also does not utilize abundant inter device communication opportunities in many scenarios. This paper proposes the human network (HUNET), a network architecture that enables information sharing between mobile devices through direct interdevice communication. We design B-SUB, an interest-driven information sharing system for HUNETs. In B-SUB, content and user interests are described by tags, which are human-readable strings that are designated by users. An experiment is performed to demonstrate the effectiveness of this tag-based content description method. To facilitate efficient data dissemination, we invent the Temporal Counting Bloom filter (TCBF) to encode tags, which also reduces the overhead of content routing. Comprehensive theoretical analyses on the parameter tuning of B-SUB are presented and verify B-SUB’s ability to work efficiently under various network conditions. We then extend B-SUB’s routing scheme to provide a stronger privacy guarantee. Extensive real-world trace-driven simulations are performed to evaluate the performance of BSUB, and the results demonstrate its efficiency and usefulness.

The Design and Evaluation of An Information Sharing System for Human Networks

With fast-growing consumer demands and rapidly-developing mobile technologies, portable mobile devices are becoming a necessity of our daily lives. However, existing mobile devices rely on the wireless infrastructure to access Internet services provided by central application providers. This architecture is inefficient in many situations and also does not utilize abundant interdevice communication opportunities in many scenarios. This paper proposes the human network (HUNET), a network architecture that enables information sharing between mobile devices through direct interdevice communication. We design B-SUB, an interest-driven information sharing system for HUNETs. In B-SUB, content and user interests are described by tags, which are human-readable strings that are designated by users. An experiment is performed to demonstrate the effectiveness of this tag-based content description method. To facilitate efficient data dissemination, we invent the Temporal Counting Bloom filter (TCBF) to encode tags, which also reduces the overhead of content routing. Comprehensive theoretical analyses on the parameter tuning of B-SUB are presented and verify B-SUB's ability to work efficiently under various network conditions. We then extend B-SUB's routing scheme to provide a stronger privacy guarantee. Extensive real-world trace-driven simulations are performed to evaluate the performance of B-SUB, and the results demonstrate its efficiency and usefulness.

Preface for the special issue of MTAP following CBMI 2011

Since the late 90s the Content-BasedMultimedia Indexing (CBMI) has become one of themore active research topics, joining efforts in the computer science (e.g., data mining, information retrieval) and signal processing (e.g., digital image and audio analysis) communities. Since the very beginning, the CBMI community has seen the evolution of research both from a methodological point of view and also from an application point of view. The proliferation of machine-learning approaches in supervised and unsupervised frameworks have witnessed to deal with the phenomenon of large scale in multimedia indexing and retrieval tasks. Content description methods have to make a progress in view of new application areas such as medical images and social media retrieval. Information fusion for efficient mining of multimedia content and also the classical task of information abstraction have tremendous opportunities when applied on multimedia data. The evaluation metrics also will have to progress in the nearest future exploiting the insights gained from the user studies in the field. The International Workshop on Content-based Multimedia Indexing has been established as one of the most important forums for the presentation of the newest ideas and original research results in the field. Following the eight successful previous events of CBMI (Toulouse 1999, Brescia 2001, Rennes 2003, Riga 2005, Bordeaux 2007, London 2008, Chania 2009 and Grenoble 2010), the Video Processing and Understanding Lab (VPULab) and the Information Retrieval Group (IRG) at Universidad Autonoma de Madrid organized the CBMI’2011 event in Madrid from 13 to 15 June 2011. Multimed Tools Appl (2013) 62:1–4 DOI 10.1007/s11042-011-0927-6

An Image Content Description Technique for the Inspection of Specular Objects

This paper proposed an image content description method within the context of specular surface inspection. Such a method is based on a preliminary research concerning the generation of specific stripe patterns for the visual enhancement of defective surface parts of cylindrical specular objects. The goal of this paper is to address the stripe pattern interpretation within a general approach. For this purpose, different pattern recognition processes, consisting not only of the combination of different image segmentation, feature retrieval, and classification, but also of feature combination and selection, will be considered. Three top-down and one bottom-up approaches are evaluated for retrieving the most appropriate feature sets in terms of highest classification rates. It will be demonstrated that following a combination and appropriate selection of these feature sets, even better rates can be reached. With only half of the initial features, an increase of more than 2% is observable.

Face Modeling and Animation Language for MPEG-4 XMT Framework

This paper proposes FML, an XML-based face modeling and animation language. FML provides a structured content description method for multimedia presentations based on face animation. The language can be used as direct input to compatible players, or be compiled within MPEG-4 XMT framework to create MPEG-4 presentations. The language allows parallel and sequential action description, decision-making and dynamic event-based scenarios, model configuration, and behavioral template definition. Facial actions include talking, expressions, head movements, and low-level MPEG-4 FAPs. The ShowFace and iFACE animation frameworks are also reviewed as example FML-based animation systems.