Fischman, Sébastien Pérez-Anker, Javiera Tognetti, Linda Di Naro, Angelo Suppa, Mariano Cinotti, Elisa Viel, Théo Monnier, Jilliana Rubegni, Pietro del Marmol, Véronique
...
Published in
Scientific Reports
Diagnosis based on histopathology for skin cancer detection is today’s gold standard and relies on the presence or absence of biomarkers and cellular atypia. However it suffers drawbacks: it requires a strong expertise and is time-consuming. Moreover the notion of atypia or dysplasia of the visible cells used for diagnosis is very subjective, with ...
De Lemos Esteves, Frédéric Schyns, Michael Jacquemin, Alexis Fettweis, Xavier Doutreloup, Sébastien
Les MOOCs (Massive Open Online Course) sont des cours en ligne scénarisés et composés de capsules vidéo-pédagogiques, d’activités pédagogiques variées, de ressources complémentaires ou encore de forums de discussions (Downes, 2008). Ces cours sont proposés gratuitement et massivement sur la plateforme FUN MOOC (https://www.fun-mooc.fr/) mais ils so...
Kharroubi, Abderrazzaq
The presentation presents the point clouds as remote sensing data for change detection. Indeed, the earth's surface is continuously changing either rapidly or slowly depending on the event. Each of these events has a defined magnitude, velocity, and frequency which determines whether the change is short-term or long-term. By reading this presentati...
Vandeghen, Renaud Cioppa, Anthony Van Droogenbroeck, Marc
We propose a novel semi-supervised learning method for leveraging unlabeled data by generating pseudo labels with a teacher-student approach. We also introduce three loss parametrizations to introduce doubt in the pseudo labels based on their confidence scores. Finally, we show that our method allows to improve detection performance for satellite i...
Fischer, Florian Bachinski, Miroslav Klar, Markus Fleig, Arthur Müller, Jörg
Published in
Scientific Reports
Among the infinite number of possible movements that can be produced, humans are commonly assumed to choose those that optimize criteria such as minimizing movement time, subject to certain movement constraints like signal-dependent and constant motor noise. While so far these assumptions have only been evaluated for simplified point-mass or planar...
Hansen, Damien Pickford, Susan
Nkamla Penka, Jean Bertin Mahmoudi, Saïd Debauche, Olivier
Agriculture 4.0 is a domain of IoT in full growth which produces large amounts of data from machines, robots, and sensors networks. This data must be processed very quickly, especially for the systems that need to make real-time decisions. The Kappa architecture provides a way to process Agriculture 4.0 data at high speed in the cloud, and thus mee...
Tadrist, Nassima Debauche, Olivier Mahmoudi, Saïd Guttadauria, Adriano
Recent floods, have shown that classical monitoring system of rivers are no longer suitable because other phenomena such insufficient capacity and / or obstruction of the drainage networks, modification of cropping practices and rotations, increase in the size of plots linked to land consolidation, urbanization of flood-prone areas, etc. The combin...
Vandeghen, Renaud Cioppa, Anthony Van Droogenbroeck, Marc
Accurate player and ball detection has become increasingly important in recent years for sport analytics. As most state-of-the-art methods rely on training deep learning networks in a supervised fashion, they require huge amounts of annotated data, which are rarely available. In this paper, we present a novel generic semi-supervised method to train...
Miftari, Bardhyl Berger, Mathias Djelassi, Hatim Ernst, Damien
The Graph-Based Optimization Modeling Language (GBOML) is a modeling language for mathematical programming enabling the easy implementation of a broad class of structured mixed-integer linear programs typically found in applications ranging from energy system planning to supply chain management. More precisely, the language is particularly well-sui...