Advances in multimedia analysis, enables access and indexing of huge and rich video content. However, delivering this heterogeneous content remains a challenge. We proposes the use of multiplex networks to combine both textual and visual semantic cues extracted from TV news videos for interactive search refinement on heterogeneous data. After preprocessing, the indexed videos can be queried, for which result space is summarized in a visual cloud helping query refinement with contextual information by combining visual and textual cues. We leverage on properties of multiplex networks to extract a hierarchy of layers, which not only contextualizes results, but also guides the construction of our visual cloud, and its interactions.

One task key to the analysis of large multimedia archive over time is to dynamically monitor the activity of concepts and entities with their interactions. This is helpful to analyze threads of topics over news archives (how stories unfold), or to monitor evolutions and development of social groups. Dynamic graph modeling is a powerful tool to capture these interactions over time, while visualization and finding communities still remain difficult, especially with a high density of links. We propose to extract the backbone of dynamic graphs in order to ease community detection and guide the exploration of trends evolution. Through the graph structure, we interactively coordinate node-link diagrams, Sankey diagrams, time series, and animations in order to extract patterns and follow community behavior. We illustrate our system with the exploration of the role of soccer in 6 years of TV/radio magazines in France, and the role of North Korea in about 10 years of Japanese news.

Multimedia data is by nature heterogeneous, conveying semantic information through multiple cues. Text analysis of closed captions already brought us understanding of the spoken information. Today's advances in computer vision now enable us to look for relevant semantic information from the visual content of real-world archives. Combining these two levels of extracted information to make sense of an archive still remains a challenge. Multiplex net- works, which model multiple families of interactions in a graph, can capture and combine both sources of semantics. We can leverage on these objects to extract hierarchies and integrate them in an interactive heterogeneous "visual cloud". Inspired by word clouds, these clouds allow to grasp visual and textual semantic information captured from a multimedia collection all at once. The interaction then enables direct access to the relevant video. We demonstrate our system with the exploration of a Japanese news archive.

The analysis and exploration of a social network depends on the type of relations at play. Borgatti had proposed a type taxonomy organizing relations in four possible categories. Homophily (similarity) relationships form an important category where relations occur when entities of the network link whenever they exhibit similar behaviors. Examples are networks of coauthor, where homophily between two persons follows from co-authorship; or network of actors having played under the supervision of the same movie director, for instance. Homophily is often embodied through a bipartite network where entities of a given type A (authors, movie directors) connect through entities of a different type B (papers, actors). A common strategy is then to project this bipartite graph onto a single-type network with entities of a same type A, possibly weighting edges based on how the type A entities interact with the type B entities underlying the edge. The resulting single-type network can then be studied using standard techniques such as community detection using edge density, or the computation of various centrality indices. This paper revisits this type of approach and introduces a homogeneity measure inspired form past work by Burt. Two entities of type B interact when they both induce a same edge between two entities of type A. The homogeneity of a subgroup thus depends on how intensely and how equally interactions occur between entities of type B giving rise to the subgroup. The measure thus differentiates between subgroups of type A exhibiting similar topologies depending on the interaction patterns of the underlying entities of type B. The method is validated using two widely used datasets. A first example looks at authors of the IEEE InfoVis Conference (InfoVis 2007 Contest). A second example looks at homophily relations between movie actors that have played under the direction of a same director (IMDB).

Who said What, Where and How? How are images, video and stories spreading out? Who produces the information? OTMedia addresses these questions by collecting, enriching and analysing continuously more than 1500 streams of French media from TV Radio, Web, AFP, and Twitter. Two studies on media produced by end users with the OTMedia framework are presented.

Exploring document collections remains a focus of research. This task can be tackled using various techniques, typically ranking documents according to a relevance index or grouping documents based on various clustering algorithms. The task complexity produces results of varying quality that inevitably carry noise. Users must be careful when interpreting document relevance or groupings. We address this problem by computing cohesion measures for a group of documents confirming/infirming whether it can be trusted to form a semantically cohesive unit. The index is inspired from past work in social network analysis (SNA) and illustrates how document exploration can benefit from SNA techniques.

This paper presents a visual-based media event detection system based on the automatic discovery of the most circulated images across the main news media (news websites, press agencies, TV news and newspapers). Its main originality is to rely on the transmedia contextual information to denoise the raw visual detections and consequently focus on the most salient transmedia events.

This paper presents 3 interfaces to access video contents. The stream explorer allows to explore and to segment video streams. The video explorer shows a synthetic view of structured TV programmes. The collection explorer proposes cartographies of large video collections. Based on visual and textual automatic processing, proximities and redundancies are analyzed, allowing the emergence of different levels of structure. This is made possible thanks to the volume of data considered: 7 channels during 100 days, ie 16000 hours or 20 Millions key frames. These three tools allow efficient exploration of video contents at different levels of interest: image, shot and sequence, programme and collection level.

We present an interactive visualization, called table of video contents (TOVC), for browsing structured TV programs such as news, magazines or sports. In these telecasts, getting a good segmentation can be very time-consuming, especially in an annotating context. This visualization, connected with a classical media player, offers a very handy video browser. This system allows a global overview by showing the temporal structure and by giving some semantic information. The drawn structure enables a non linear video access by suggesting relevant frames. The TOVC is created from a graphic framework designed for computing similarities on visual content, and displaying the associated proximities in a 2D map with graph representation. TOVC is one of its first applications and shows interesting capabilities.

This paper describes an augmentation to the spline concept to account for topological relations between different spline curves. These topological relations include incidence relations, constraining the extremes of spline curves to other spline curves, and also more general geometric relations, for example, involving the tangents of spline curves in their extremes. To maintain these incidence relations, some spline curves may have to be transformed (translated, rotated, scaled), or even deformed (i.e., the shape of the curve may change) as a result of modifying other spline curves. A data structure and algorithms are given to implement the propagation of these transformations and deformations. Based on the augmented spline concept, to be called sticky splines , both a script system to represent spline structures and an interactive system for editing drawings while automatically, maintaining their topological structure are presented.

The use of Monte-Carlo integration together with stochastic sampling is very useful for dealing with the scattering phenomena that occur in the propagation and reflection of light. In this paper, these techniques have been applied to the implementation of a physicsbased global illumination model. The theoretical basis of this approach is presented briefly and various applications to realistic image rendering are then described. This concerns the rendering of penumbra and scattered reflection effects, antialiasing and accurate color modelling through spectral integration. For all these applications, both theoretical and implementation aspects are developed and it is shown that stochastic techniques can provide very simple and efficient algorithms.