-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathcurrentProjects.html
146 lines (143 loc) · 11.6 KB
/
currentProjects.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
<div class="row">
<div class="col-sm-4">
<div class="panel-body">
<div class="list-group" id="list-tab" role="tablist">
<a class="list-group-item list-group-item-action active" id="list-home-list" data-toggle="list" href="#projULTRALEARN"
role="tab" aria-controls="home">ANR Project JCJC Ultra-Learn </a>
<a class="list-group-item list-group-item-action" id="list-home-list" data-toggle="list" href="#projRVESSELX"
role="tab" aria-controls="home">ANR Project R-VESSEL-X </a>
<a class="list-group-item list-group-item-action " id="list-home-list" data-toggle="list" href="#projPARADIS"
role="tab" aria-controls="home">ANR Project JCJC PARADIS</a>
</div>
</div>
</div>
<div class="col-sm-8">
<div class="tab-content">
<!-- Begin projULTRALEARN -->
<div class="tab-pane proj fade show active " id="projULTRALEARN" role="tabpanel" aria-labelledby="list-home-list">
<h5>Ultra-Learn: Supervised Ultrametric Learning</h5>
<ul>
<li><b>Project Type</b>: ANR (French founding)</li>
<li><b>Project coordinator</b>: <b>Benjamin Perret</b></li>
<li><b>Period:</b> 04/2021 - 2025 (48 Months)</li>
<li><b>Hosting institutions:</b> <a href="http://ligm.u-pem.fr/" target="_blank">LIGM</a>,<a href="https://www.esiee.fr/" target="_blank">ESIEE Paris</a>,<a href="https://www.univ-gustave-eiffel.fr/en/" target="_blank">Université Gustave Eiffel</a></li>
<li><b>Members:</b>
<ul>
<li><a href="https://perso.esiee.fr/~perretb/" target="_blank">Benjamin Perret (PI)</a></li>
<li><a href="https://perso.esiee.fr/~chierchg/" target="_blank">Giovanni Chierchia</a></li>
<li><a href="https://perso.esiee.fr/~coustyj/" target="_blank">Jean Cousty</a></li>
<li><a href="http://www.laurentnajman.org/" target="_blank">Laurent Najman</a></li>
</ul>
<li><b>Website </b>: from <a href="https://anr.fr/Projet-ANR-20-CE23-0019">anr</a> and <a href="https://perretb.github.io/UltraLearn/">official project page</a></li>
<li><b>Abstract</b>:
An ultrametric is a dual representation of a hierarchical clustering which is a classical machine learning method used when class
labels are not known in advance. This approach is motivated theoretically by the fact that complex data such as social networks,
power grids, actor networks, computer networks, cortical brain networks, language networks, and so forth, naturally show
hierarchical organizations. Being able to construct high quality hierarchical representations of those data is thus an important step
toward their analysis and their understanding. While hierarchical clustering is generally regarded as an unsupervised method, the
use of supervision will greatly improve the quality of hierarchical methods and thus their capacity at automatically discovering
meaningful organization in data. Therefore, this project aims at developing scalable methods for supervised learning of hierarchical
clustering.
</li>
</ul>
</div>
<!-- End projULTRALEARN -->
<!-- Begin projRVESSELX -->
<div class="tab-pane proj fade show" id="projRVESSELX" role="tabpanel" aria-labelledby="list-home-list">
<h5>R-VESSEL-X: Robust vascular network extraction and understanding within hepatic biomedical images</h5>
<ul>
<li><b>Project Type</b>: ANR (French founding)</li>
<li><b>Project coordinator</b>: <b>Antoine Vacavant (Institut Pascal)</b></li>
<li><b>Period:</b> 01/2019 - 2022 (48 Months)</li>
<li><b>Partners:</b>
<ul>
<li> Kitware, SAS (Lyon): partener coordinator: <b>Julien Jomier</b>, participants: Laurenn Lam </li>
<li> Institut Pascal: partener coordinator: <b>Antoine Vacavant</b>, participants: Hugo Rositi, Manuel Grand Brochier, Nicolas Beaurepaire, Armand Abergel, Benoît Magnin, Pascal Charbot </li>
<li>CReSTIC, Reims, Lyon: partener coordinator: <b>Nicolas Passat</b>, participants: Hervé Deleau, Stéphanie Salmon, Devy Jérôme, Christophe Portefaux, Hugues Talbot.
</li>
<li> LIRIS, Lyon (France): partener coordinator: <b>Bertrand Kerautret</b>, participants: Laure Tougne, Serge Miguet,
Carlos F. Crispin-Junior including LORIA Nancy (France) members: Isabelle Debled-Rennesson, participants:
Philippe Even, Phuc Ngo </li>
</ul>
<li><b>Website </b>: from <a href="https://anr.fr/Project-ANR-18-CE45-0018">anr</a> and official project page <a href="http://tgi.ip.uca.fr/r-vessel-x/">here</a></li>
<li><b>Abstract</b>:
Cardiovascular diseases and other blood vessels disorders increase in the world-wide scale and in particular
in Occident. The evolution of computer science in researches investigating vascular networks has raised the
interest in numerical reconstructing and understanding of those complex tree-like structures. The R-VESSEL-X
project proposes original and robust developments of image analysis and machine learning algorithms
integrating strong mathematical frameworks, e.g. digital geometry and topology, mathematical morphology, or
graphs for reconstructing vessels of the liver beyond medical image content. Another objective of R-VESSEL-X
is to diffuse research works in an open-source way, with the developments of plug-ins compatible with the
ITK and VTK libraries largely popularized by the KITWARE company. This project will also include benchmarks
composed of images, associated ground-truth and quality metrics, so that researchers and engineers evaluate
their novel contributions. The consortium of R-VESSEL-X is composed of the following laboratories: Institut
Pascal (coordinator, Le Puy-en-Velay/Clermont-Ferrand), LIRIS (Lyon), CReSTIC (Reims), working together with
the KITWARE company (Lyon). This is a highly pluridisciplinary group composed of researchers in computer
science-related topics (biomedical image processing, numerical simulation and analysis), applied mathematics
(digital geometry and topology, mathematical morphology), working with medical doctors (radiologists,
hepatologists) and young researchers and developers enrolled for the project.
</li>
</ul>
</div>
<!-- End projRVESSELX -->
<!-- Begin projPARADIS -->
<div class="tab-pane proj fade show fade show " id="projPARADIS" role="tabpanel" aria-labelledby="list-home-list">
<h5>PARADIS: PARameter-free Analysis of DIgital Surfaces</h5>
<ul>
<li><b>Project Type</b>: ANR JCJC (French founding)</li>
<li><b>Project coordinator</b>: <b>Tristan Roussillon (LIRIS, Lyon)</b></li>
<li><b>Period:</b> December 2018 - December 2022 (42 Months)</li>
<li><b>Partners:</b>
<ul>
<li> LIRIS, Lyon, France </li>
<li> Collaborators: David Coeurjolly (CNRS, LIRIS), Bertrand Kerautret (Lyon2, LIRIS), Sébastien Labbé
(LABRI, Bordeaux, France), Jacques-Olivier Lachaud (Savoie Mont Blanc, LAMA)
</li>
</ul>
<li><b>Website </b>: from <a
href="https://anr.fr/en/funded-projects-and-impact/funded-projects/project/funded/project/b2d9d3668f92a3b9fbbf7866072501ef-a38ced7290/?tx_anrprojects_funded%5Bcontroller%5D=Funded&cHash=1d77a3443d38ffb7079506dc8cea294b">anr</a>
and official project page <a href="https://perso.liris.cnrs.fr/tristan.roussillon/paradis.html">here</a>
</li>
<li><b>Abstract</b>:
In numerous fields (e.g., material sciences, medical imaging), non-invasive acquisition devices such as
magnetic resonance, X-ray tomography or microtomography are required for observation, measurements or
diagnostic aids. These acquisition devices usually generate volumic data, i.e., 3D images, composed of
regularly spaced data in a cuboidal domain. 3D volumes come from the segmentation of such 3D images. They
can also be generated from scientific modeling because numerous simulation schemes rely on the regularity
of the data support.
PARADIS is a project about geometry of 3D volume boundaries, called digital surfaces. Keeping the digital
nature of the data is an advantage to do integer-only and exact computations, to perform constructive
solid geometry operations, or to use efficient spatial data structures. A drawback is its poor geometry: a
digital surface is only made up of quadrangular surface elements whose normal vector is parallel to one
axis, whatever the resolution. Many high-level tasks in computer graphics, computer vision and 3D image
analysis require a richer geometry, e.g., rendering, surface deformation for physical simulation or
tracking, precise geometric measurements. To perform relevant geometric tasks and to benefit from the
above-mentioned advantages in the same time, one need to enhance the geometry of digital surfaces by
estimating extra information for each surface element. This project focuses on estimations of local and
first-order geometric quantities such as normal vector direction. It aims at providing accurate and
parameter-free estimators based on a surface patch with adaptive size around each surface element. Since
we are looking for first-order estimations, the patch will be typically a piece of digital plane that
locally fits the digital surface.
A challenge is to cover the whole digital surface by maximal pieces of digital plane. Such a cover not
only provides a normal vector field, but may also provide, if computed for several subsampled versions of
the input 3D volume, a way of determining the scale at which noise is unlikely: a high number of very
small digital plane segments indicate noise, whereas smooth parts are decomposed into a smaller set of
larger segments. What is challenging is that there is a combinatorial explosion of maximal pieces of
digital plane and that among them, not all are tangent to the digital surface.
An opportunity to make a breakthrough regarding this issue is to consider the recent development of
plane-probing algorithms, proposed by the principal investigator and his collaborators. These algorithms
allow to decide on-the-fly how to probe the digital surface and make grow a digital plane segment, which
is tangent by construction. The growth direction is given by both arithmetic and geometric properties.
PARADIS aims at analyzing digital surfaces with the help of plane-probing algorithms.
We expect a positive impact in computer graphics, computer vision and 3D image analysis because the
above-mentioned high-level tasks and many others, such as primitive extraction or scene understanding,
rely on the quality of normal estimation. In addition, since many 3D images may be degraded with noise,
especially in medical imaging, noise detection is a crucial task, which may become a key step of 3D image
processing pipelines.
</li>
</ul>
</div>
<!-- End projPARADIS -->
</div>
</div>
</div>