Catálogo de publicaciones - libros

We present COHIBIT, an edutainment exhibit for theme parks in an ambient intelligence environment. It combines ultimate robustness and simplicity with creativity and fun. The visitors can use instrumented 3D puzzle pieces to assemble a car. The key idea of our edutainment framework is that all actions of a visitor are tracked and commented by two life-like guides. Visitors get the feeling that the anthropomorphic characters observe, follow and understand their actions and provide guidance and motivation for them. Our mixed-reality installation provides a (via the graspable car pieces), (via the coordinated speech, gestures and body language of the virtual character team) and (via the large-size projection of the life-like characters) experience for a single visitor or a group of visitors. The paper describes the context-aware behavior of the virtual guides, the domain modeling and context classification as well as the event recognition in the instrumented environment.

The Nintendo DS is a hand held game computer that includes a small sketch pad as one of it input modalities. We discuss the possibilities for recognition of simple line drawing on this device, with focus of attention on robustness and real-time behavior. The results of our experiments show that with devices that are now becoming available in the consumer market, effective image recognition is possible, provided a clear application domain is selected. In our case, this domain was the usage of simple images as input modality for computer games that are typical for small hand held devices.

In this paper, we present a game-like scenario that is based on a model of social group dynamics inspired by theories from the social sciences. The model is augmented by a model of proxemics that simulates the role of distance and spatial orientation in human-human communication. By means of proxemics, a group of human participants may signal other humans whether they welcome new group members to join or not. In this paper, we describe the results of an experiment we conducted to shed light on the question of how humans respond to such cues when shown by virtual humans.

This paper presents a virtual dancer that is able to dance to the beat of music coming in through the microphone and to motion beats detected in the video stream of a human dancer. In the current version its moves are generated from a lexicon that was derived manually from the analysis of the video clips of nine rap songs of different rappers. The system also allows for adaptation of the moves in the lexicon on the basis of style parameters.

In this paper we investigate the role of user emotions in human-machine goal-oriented conversations. There has been a growing interest in predicting emotions from acted and non-acted spontaneous speech. Much of the research work has gone in determining what are the labels and emotion prediction accuracy. In this paper we evaluate the of user emotional state towards a computational model of emotion processing. We consider a binary representation of emotions (positive vs. negative) in the context of a goal-driven conversational system. For each human-machine interaction we acquire the temporal emotion sequence going from the initial to the final conversational state. These traces are used as features to characterize the user state dynamics. We ground the emotion traces by associating its patterns to dialog strategies and their effectiveness. In order to quantify the of emotion indicators, we evaluate their predictions in terms of speech recognition and spoken language understanding errors as well as task success or failure. We report results on the 11.5 dialog corpus samples from the corpus.

”Fluids” is an interactive and immersive mixed reality art installation that explores the relation of intimacy between reality and virtuality. We live in two different but connected worlds: our physical environment and virtual space. In this paper we discuss how we integrated them by using water and air as interfaces. We also discuss how we designed mechanisms for natural and subtle navigation between and within the different environments of the piece, how we designed the environments and the installation so as to take advantage of the low cost alternatives that are available today.

Location-based games introduce an element that is missing in interactive console games: movements of players involving locomotion and thereby the physical effort characteristic of any sportive activity. The paper explores how to design location-based games combining locomotion with strategic reasoning by using classical board games as templates. It is shown that the straightforward approach to “spatialize” such games fails. A generic approach to spatialization is presented and described within a conceptual framework that defines a large class of geogames. The framework is complemented by a software tool allowing the game designer to find the critical parameter values which determine the game’s balance of reasoning skills and motoric skills. In order to illustrate the design method, a location-based version of the game TicTacToe is defined and analyzed.

Here we present a model of a subtype of one-line jokes (not puns) that describes the relationship between the (part of the set-up) and the (often called the punchline). This relationship is at the heart of what makes this common type of joke humorous. We have implemented this model in a system, DisS (Disjunctor Selector), which, given a joke set-up, can select the best disjunctor from a list of alternatives. DisS agrees with human judges on the best disjunctor for one typical joke, and we are currently testing it on other jokes of the same sub-type.

We have studied the use of a public screen integrated to a mobile multiplayer game in order to create a new kind of user experience. In the user evaluations, the game FirstStrike was tested by eight user groups, each containing four players. The evaluations showed that communication between the players increased with the usage of the public display and alliances were built. It was also found that the possibility to identify the players by adding the players’ photographs into the shared display makes the game more personal. We believe that this new way of communication is a result of using the shared public screen in a mobile multiplayer game.

Augmented Reality technology permits the concurrent interaction with the real environment and computer-generated virtual objects, thus making it an interesting technology for developing educational applications that allows manipulation and visualization. The work described extends the traditional concept of book with rendered graphics to help children understand fundamentals of the theory of colours. A three-dimensional virtual chameleon shows children how, from the combination of primary colours, it is possible to get secondary colours and viceversa. The chameleon responds to children’s actions changing appearance according to the colours of the surroundings. Our tangible interface becomes an innovative teaching tool conceived for supporting school learning methods, where the child can learn by playing with the virtual character, turning over the pages of the book and manipulating the movable parts. The main scientific contribution of this work is in showing what the use of augmented reality-based interfaces can bring to improve existing learning methods.