We present some results of the research-action project IIWAC (Innovative Interactions Weaving by Artistic Creation, or Tissage d'Interactions Sociales Innovantes par la Creation Artistique - TISICA in French) in which artists worked with dependent elderly people suffering from Alzheimer's disease, residing in a specialized service of the hospital. The artists led workshops based on musical and sound play, pictorial creation and dance, with the integration of connected objects dedicated to the sonification of the gestures. For this integration, we used two functional prototypes linking gesture and music: connected smartwatch and ring. The testimonials of nursing staff members explicit how the elders regain dignity through this project, discovering new social skills, and sharing deep life experiences with other hospital residents and caregivers.

This paper describes LeBonGeste, a device consisting of a basketball basket equipped with sensors, actuators and computers to create a so called augmented basket. It is used to train beginner players to the free throw by helping them to achieve the right stance. Thanks to the Kinect depth-sensor, the system delivers realtime feedback to the player. LeBonGeste has been showcased two times indoor and outdoor using different communication channels for feedback: visual (screen or LED strips) and audio. Developed as an interactive installation for training by entertaining, the system will be tested soon on basketball courts with experienced players.

The SIFT feature extractor was introduced by Lowe in 1999. This algorithm provides invariant features and the corresponding local descriptors. The descriptors are then used in the image matching process. We propose an overview of this algorithm: the methodology and the tricky steps of its implementation, properties of the detector and descriptor. We analyze the structure of detected features. We finally compare our implementation to others, including the Lowe’s.

We present an approach for 3D reconstruction of a city model over the time from a collection of old postcards of the city of Reims. The planar structure of the buildings facades constraints the dense reconstruction of the city. We use a feature matching technique while proposing the registration of facades in the images and there use for the reasoning about there visibility in the images. This system is semi-automatic, it requires a user control in the complicated case where no matches are found to link an image to the dataset. The image data set is sparse and the urban space evolves over time.

We propose an experimental design for the comparison of state-of-the art feature detector-descriptor combination. Our aim is to rank potential detector-descriptor that best performs in our project. We deal with disparate images that represent building evolution of the city of Rheims over the time. We obtained promising results for matching buildings that evolve temporally.

This paper proposes new uses of cloud computing to facilitate in-class learning processes. We present them through the case study of a Human-Computer Interaction course. Google Drive tools were used for their collaborative abilities, in order to reach different pedagogical aims. These goals are divided in human and technical aspects. On the relational side, we find the inclusion process at the beginning of the course, the need to collaborate to achieve a common objective, the practice of one-to-one and many-to-many collaboration, the ability to present one's work to other people and being evaluated, and the ability to evaluate another's work. On the technical side, the students learned the use of a collaborative tool, the fundamental theoretical laws of human-computer interaction, and their practical implementation.

We present the first step of the Rheims city (France) 3D reconstruction based on old postcards. Rheims endured drastic destruction during the First and then the Second World Wars. Old postcards, embodying the period from the beginning to the end of the 20th century, establish a testimony of the evolution in time and space of the most important places and monuments in the city. We aim to localize, render and visualize in 3D the buildings of this city, by using this sparse dataset and incomplete cadastral surveys. Image matching is initially required for locating a building in the city space. In this paper we propose a performance comparison of state-of-the-art detector-descriptor couples. We focus on their robustness and their transformations invariance when low resolution data is subject to temporal and space modifications.

We address the problem of the 3D reconstruction of Rheims (France) using sparse data. We consider an old postcard dataset embodying the period from the beginning to the middle of the 20th century. Our reconstruction have to take into account spatial and temporal aspects that are implicitly characterizing our dataset. Indeed Rheims was destroyed during the first World War, we dispose of postcards taken before, during and after the war. The first steps of our work consists in determining the best way to match different views of the same place. This study reports on performance comparison of state-of-the-art couple of detectors-descriptors that are most robust and reliable in the context of our application.

As fMRI data is high dimensional, applications like connectivity studies, normalisation or multivariate analyses, need to reduce data dimension while minimising the loss of functional information. In our study we use connectivity profiles as a new functional feature to aggregate voxels into clusters. This offers two major advantages in comparison with the current clustering methods. It allows the analyst to deal with the spatial correlation of noise problem, that can lead to bad mergings in the functional domain, and it is based on the whole data independently of a priori information like the General Linear Model (GLM) regressors. We validated that the resulting clusters form a partition of the data in homogeneous regions according to both spatial and functional criteria.

Les annees 90 ont ete le theâtre d'une proliferation des technologies numeriques au sein de toutes les strates de l'activite humaine. La numerisation de l'information (sons, images fixes ou animees) a engendre des modifications profondes au niveau de sa diffusion, de son archivage et de sa consultation. La communaute artistique, pressentant cette rupture dans le rapport que nous entretenons a l'information, s'est tres tot interessee aux potentialites des technologies numeriques. Apport des technologies numeriques a la creation artistique En effet, des oeuvres, que l'on peut qualifier de numeriques a partir du moment ou ces technologies interviennent dans le processus artistique, existent en tres grands nombre et sous des formes tres variees. De plus, on constate que les technologies numeriques peuvent intervenir a differentes etapes du processus creatif. Dans ces conditions, on peut etre tente de definir des categories d'artistes (ou d'oeuvres d'art) basees sur le moment ou interviennent les technologies numeriques :

Thanks to the use of autofluorescent proteins, cell biologists are now able to observe and study the evolution of the volumetric content of live cells through time sequences of 3D images (also called 4D images). We present in this paper 4DReAM a tracking framework based on a deformable model, and how we apply it in cell biology to understand, describe, explain, in addition to interactively visualize and analyze the evolution of structures enclosed one in another, at different levels in the cell: proteins, nucleoli and nucleus. This deformable model, defined as a triangulated mesh, can automatically adapt its topology to fit the splits and merges of the biological objects. The usefulness of 4DReAM is illustrated on experimental results from different 4D images acquired by confocal microscopy.

We present an original approach of virtual sculpting incarnated in a unified space-time interactive modeler named Splimo (for Spline Modeler), which is specifically designed for the creation of tubular spline works. Oriented towards artistic creation, it allows the realization of virtual transcriptions of real art pieces, as well as the creation of animated virtual sculptures, presented in an exhibition mode. Drawing curves in the virtual world is a way to be free of the constraints of the real sculptures, that's why we call it "ideal". Splimo may however also be used as a virtual prototyping tool to experiment new shapes to be later designed in the real world.

We present a method to automatically extract the evolution of the cell envelope in 4D confocal images. Our method is based on 4DReAM, a tracking system consisting of a previously presented deformable surface model, which can change its topology. The process consists in attracting the model for each volume of a 4D series towards an iso-surface of interest, and towards the image gradients. We then statistically estimate the characteristics of the cell surface on the nodes of the model, and reconstruct it. We show detailed results on the segmentation of the cell envelope during the mitosis.

We present in this paper the hierarchical extension of 4DReAM, a tracking framework based on a deformable model, and how we apply it in cell biology, where the use of auto fluorescent proteins and confocal microscopy enables the acquisition of the volumetric content of live cells through time sequences of 3D images (also called 4D images). Thanks to a hierarchical paradigm, 4DReAM allows interactive visualisation and analysis of the evolution of structures enclosed one in another, at different levels in the cell, such as nucleus, nucleoli, and proteins. The usefulness of our approach is illustrated on experimental results from confocal 4D images, which bring new light to the space-time reorganisation of intra-cell structures.

Recent developments in volume visualization using standard graphics hardware provide an effective and interactive way to understand and interpret the data. Mainly based on 3d texture mapping, these hardware-accelerated visualization systems often use a cell-projection method based on a tetrahedral decomposition of volumes usually sampled as a regular lattice. On the contrary, the method we address in this paper considers the slicing problem as a restricted solution of the marching cubes algorithm [1,2]. Our solution is thus simple, elegant and fast. The nature of the intersection polygons provides us with the opportunity to retain only 4 of the 15 canonical configurations defined by Lorensen and Cline and to propose a special look-up table.

In this paper, we present an integrated approach to visualize and quantitatively analyze spatial processes in time sequences of 3D images, also called 4D images. We describe how the use of a deformable surface model which may change its topology, is relevant to solve the difficulties that arise in this field. The objects embedded in the volumes are extracted and organized in a way to establish the correspondences needed by the temporal evolution. Furthermore, the data between two successive images is interpolated to reconstruct the objects evolution. Descriptors of the evolution are drawn up and computed, as a means to quantify the phenomenon. They are added to the model, leading to nDimensional data, which is structured in a direct acyclic graph 7 We show that the evolution may be synthetized and represented with simple metaphors which we integrate in an interactive visualization tool.Our method is illustrated by experimental results on real biological data obtained by confocal scanning microscopy.

Display of large volumes, progressive rendering and selective refinements are some of the operations supported by multiresolution technology. In this paper, a general framework relying on the use of such techniques applied to volume data rendering is presented. Based on a pyramidal representation of data, two aspects of our work are considered. First, the decimation algorithm itself is described. The general principle consists in gradually removing nodes of a structural graph previously established while respecting constraints. A second part introduces the refinement of preliminary obtained Levels of Details (LOD). The problem is to preserve as well as possible the initial volume of studied objects. The goal is to build an interactive system of visualization for the analysis of volumetric data. The speed of treatments associated with a good visualization should enable to achieve a 3D survey of a natural object in a quasi-interactive manner. The method has been successfully applied to both synthetic and real data (medical imaging).

This chapter discusses a technique for registering three-dimensional sets of particles which are related by nonrigid deformations. These sets of particles are usually surfaces or sets of surface patches to which are assigned features such as gradient, curvature, etc. Two arguments which favored high-level feature-based approaches in the past were computational complexity and global correspondence search. The computational complexity argument assumed that if a small set of features was found and matched (say dozens of discrete features or hundreds of ridge points, as opposed to tens of thousands of surface particles), the overall complexity of the algorithm would be drastically lower. However, by using the oct-tree spline distance map, the complexity of each iterative adjustment step in our algorithm is linear in the number of sensed surface particles. The second argument in favor of discrete features is related to the first, and examines the combinatorial complexity of certain discrete matching (correspondence) algorithms. In its approach, the chapter uses a modified gradient descent which avoids combinatorial search but only finds locally optimal matches. In theory, volumetric deformation models are of a higher complexity than elastic surface models, that is, they represent the deformations over a dense region of three-dimensions, rather than over a two-dimensional manifold. However, the use of a multiresolution, coarse-to-fine strategy results in a very efficient matching algorithm. The estimation of elastic deformations in general is an ill-posed problem for which it is crucial to define some notion of minimal deformation that relates directly to the notion of smoothness.