Embedded Files
Here are some projects I have recently been working on. Those are still active research themes, I'm just waiting for new funds to come (please donate ;-), or more time for research, that would require a second life (keep yours, it's valuable!).
Fibravasc project
Fibravasc project
In the Fibravasc project (funded by the région Centre Val de Loire and in collaboration with neurosurgeons Ilyess Zemmoura and Christophe Destrieux from the iBrain lab), we have studied the vascular system of the brain. Our role in this project (my colleague Romain Raveaux and myself) consisted in studying a user interface to explore the large trees generated from the images of brain vessels (as the one shown).
Visual and interactive interfaces + optimization methods for improving tractography in the human brain (incl. the PhD of Barthélémy Serres, Marouen Sta).
Visual and interactive interfaces + optimization methods for improving tractography in the human brain (incl. the PhD of Barthélémy Serres, Marouen Sta).
See more details about the Fibratlas project (with funds from FEDER/Région Centre Val de Loire, ANR).
Ilyess Zemmoura, Barthélemy Serres, Frédéric Andersson, Laurent Barantin, Clovis Tauber, Isabelle Filipiak, Jean-Philippe Cottier, Gilles Venturini, Christophe Destrieux: FIBRASCAN: A novel method for 3D white matter tract reconstruction in MR space from cadaveric dissection. NeuroImage 103: 106-118 (2014)
Automatic or Interactive classification of Insect Images (PhD of Maxime Martineau).
Automatic or Interactive classification of Insect Images (PhD of Maxime Martineau).
This work is done in collaboration with Innophyt with several grants (Caramba Project and FEDER - Région Centre Val de Loire).
Maxime Martineau, Donatello Conte, Romain Raveaux, Ingrid Arnault, Damien Munier, Gilles Venturini: A survey on image-based insect classification. Pattern Recognition 65: 273-284 (2017).
Patent FR3017728B1.
Clustering metagenomic data with proximity graphs (PhD of Shivani Shah).
Clustering metagenomic data with proximity graphs (PhD of Shivani Shah).
This work was done in collaboration with Jacques-Henri Sublemontier (CEA) and Fatma Bouali.
Visual and interactive interfaces for preprocessing large sets of Open Data files (PhD of Paulo Da Silva Carvalho).
Visual and interactive interfaces for preprocessing large sets of Open Data files (PhD of Paulo Da Silva Carvalho).
This work was done in collaboration with the LIST.
Paulo da Silva Carvalho, Patrik Hitzelberger, Benoît Otjacques, Fatma Bouali, Gilles Venturini: Information Visualization for CSV Open Data Files Structure Analysis. IVAPP 2015: 101-108
Exploring a large database of images. The clustering step is performed by an hybrid approach that combines proximity graphs with BIRCH (PhD of Frederic Rayar).
Exploring a large database of images. The clustering step is performed by an hybrid approach that combines proximity graphs with BIRCH (PhD of Frederic Rayar).
Frédéric Rayar, Sabine Barrat, Fatma Bouali, Gilles Venturini: APoD eXplorer: Recommendation System and Interactive Exploration of a Dynamic Image Collection. IV 2016: 118-123. (Open in DBLP)
A clustering algorithm based on Delaunay triangulation (PhD of Octavio Razafindramanana).
A clustering algorithm based on Delaunay triangulation (PhD of Octavio Razafindramanana).
Octavio Razafindramanana, Gilles Venturini: Delaunay simplices pruning based clustering. ESANN 2013 (Open in DBLP)
DataTube2. This system is intended to let you explore a large set of time series.
DataTube2. This system is intended to let you explore a large set of time series.
Download and test DataTube2.
Fatma Bouali, Sébastien Devaux, Gilles Venturini: Visual mining of time series using a tubular visualization. The Visual Computer 32(1): 15-30 (2016)
In this other paper, we extended this visualization to represent logs of a parallel computation.
Sébastien Devaux, Fatma Bouali, Gilles Venturini: DataTube4log: A Visual Tool for Mining Multi-threaded Software Logs. IV 2014: 189-195
Below, two posts on YouTube, with an intermediate version of DataTube2.
A user assistant (VizAssist) to help you select and configure your data visualizations (PhD of Abdelheq Guettala)
A user assistant (VizAssist) to help you select and configure your data visualizations (PhD of Abdelheq Guettala)
Please test VizAssist and give us some feeback.
Fatma Bouali, Abdelheq Et-tahir Guettala, Gilles Venturini: VizAssist: an interactive user assistant for visual data mining. The Visual Computer 32(11): 1447-1463 (2016). (Open in DBLP)
Visual and interactive exploration of a large dataset. We used GPU programming to visualize and interact with millions of data (with computer response in less than 1 second).
Visual and interactive exploration of a large dataset. We used GPU programming to visualize and interact with millions of data (with computer response in less than 1 second).
This work was initiated several years ago with the PhD of David Da Costa (2007).
Tianyang Liu, Fatma Bouali, Gilles Venturini: On visualizing large multidimensional datasets with a multi-threaded radial approach. Distributed and Parallel Databases 34(3): 321-345 (2016). (Open in DBLP)
VR4OLAP. A 3D and immersive interface to explore Olap data cubes.
VR4OLAP. A 3D and immersive interface to explore Olap data cubes.
Sébastien Lafon, Fatma Bouali, Christiane Guinot, Gilles Venturini: On studying a 3D user interface for OLAP. Data Min. Knowl. Discov. 27(1): 4-21 (2013) (Open in DBLP)
Reorganizing OLAP hierarchies with a genetic algorithm (with the aim of improving the cube's visualization).
Reorganizing OLAP hierarchies with a genetic algorithm (with the aim of improving the cube's visualization).
Sébastien Lafon, Fatma Bouali, Christiane Guinot, Gilles Venturini: Hierarchical Reorganization of Dimensions in OLAP Visualizations. IEEE Trans. Vis. Comput. Graph. 19(11): 1833-1845 (2013) (Open in DBLP)
Building and visualizing a large proximity graph with GPU programming. We applied this methodology, called EXOD, to a set of 150k open datasets.
Building and visualizing a large proximity graph with GPU programming. We applied this methodology, called EXOD, to a set of 150k open datasets.
Tianyang Liu, Fatma Bouali, Gilles Venturini: EXOD: A tool for building and exploring a large graph of open datasets. Computers & Graphics 39: 117-130 (2014) (Open in DBLP)
Google Sites
Report abuse