Karl Daher

An exoskeleton helping patients in their physiotherapy rehabilitation is currently under development. It helps the patient with walking difficulties. Moreover, the physiotherapist would like to let the patient alone without being worried and so he can be more autonomous. The goal of this project is to develop a prototype detecting a way to stop reliably the exoskeleton. Indeed, it has to be secure for the patient to use the machine. In an emergency situation, we want to stop it reliably. The possible ways to stop the machine are compared: speech recognition, touch and head gestures detection. Speech recognition is implemented in the final prototype. The machine can also be stopped by touch with a button. Tests of recognition and user testing shows the ability of the prototype to be used in a real case.

Les différentes technologies de réalité mixte, dite XR, prennent de plus en plus d’ampleur, que ce soit dans le domaine de l’entreprise ou dans le foyer de chacun. La réalité virtuelle permet de s’immerger dans un monde virtuel à l’aide d’un casque se plaçant devant nos yeux. Cette technologie est à la fois utilisée dans le monde du divertissement, que ce soit pour des spectacles ou dans les jeux vidéo, mais également pour des applications scientifiques comme dans la médecine par exemple pour le soin des douleurs fantômes. La réalité augmentée quant à elle...¨ En liant ces technologies grâce à une connexion réseau, nous obtenons un environnement de travail collaboratif à distance et ouvrons de nombreuses opportunités sur des interactions possibles

Our empathic system is a system that provides multiple scientifically proven techniques for stress reducing and productivity increase for users in occupational health sector. The main objective of this system is to allow the users to use the provided techniques in multiple situations (i.e. car, office, remote work) in a way that almost all users can benefit from them without being limited to have some extra stuff in order to practice them. Our proposed system is divided into two main components

Over the years, online education has gradually become more prevalent in everyday practices. It is especially this past year that has shown the progress that still needs to be made in this sector. With the health crisis and the ensuing lockdowns, online courses have become an overnight necessity and a staple. Despite the improvements made to accommodate these unusual and new circumstances, online courses, whether real-time or pre-recorded, are woefully lacking in interaction. Being in the comfort of one's own home with generally no supervision it is easy for the student to succumb to the distractions of his or her environment or to pay passive and sometimes no attention to the lecture session taking place on the screen. This leads to passive and poor learning for the student who is no longer fully capable of retaining the necessary and crucial information. Thus, it is necessary and even vital to make online courses interactive to improve students’ experience on the one hand and teachers on the other. After extensive research, it was found that the integration of questions into the course sessions is the most effective way to energize the student, encourage them and capture their attention. In addition, this tool would allow the teacher to get a general overview of the students' performance and understanding of the concepts being stated.

Emotion style transfer is a recent and challenging problem in NLP. Transformer-based language models are becoming extremely powerful, so one wonders if it would be possible to leverage them to perform emotion style transfer. So far, previous work has not used transformer-based models for this task. To address this task, we fine-tune a GPT-2 model with corrupted emotional data. This will train the model to increase the emotional intensity of the input sentence. Coupled with a paraphrasing model, we develop a system capable of transferring an emotion into a paraphrase. We conducted a qualitative study with human judges, as well as a quantitative evaluation. Although the paraphrase metrics show poor performance compared to the state of the art, the transfer of emotion proved to be effective, especially for the emotions fear, sadness and disgust. Such technology can significantly facilitate the automatic creation of training sentences for natural language understanding (NLU) systems, but it can also be integrated into an emotional or empathic dialogue architecture.

All over the world, the consumption of food packaging has taken on disproportionate proportions. Although an eco-responsible trend emerged several years ago, many products found in supermarkets are still protected by single-use packaging and sometimes difficult to recycle. There is no need to explain the disastrous consequences this phenomenon can have on the environment. But the problem of packaging must also be analysed from a health point of view, since some packagings can contaminate the food they are trying to protect.

Durant l'été 2020, nous avons réalisé une série de vidéos avec la participation d’une danseuse contemporaine. Son rôle était d’illustrer différentes émotions (joie, colère, tristesse, etc.) au travers de courtes performances. En plus de ces performances, nous avons également filmé l’ombre de la danseuse derrière un écran. Le but de ces vidéos sera de pouvoir reconnaître et classifier automatiquement ces émotions. Afin de pouvoir reconnaitre les émotions des danseurs, nous avons besoin d’analyser ces vidéos.

De nos jours, les artistes ont du mal à se faire connaitre, même au sein de leur communauté. En parallèle, les jeux prennent de plus en plus de place dans la vie de tous. Enfants et adultes utilisent de plus en plus d’applications ludiques. Dans le but d’associer l’art et la technologie, ce projet va permettre de concevoir des jeux afin de mettre en avant les travaux d’un artiste local.

Durant ce projet de semestre, nous avions pour but de transformer un questionnaire actuellement sous format papier en application web. Le but de ce changement est de simplifier le suivi des patients et de permettre aux médecins de ne pas avoir à conserver une grande quantité de papier.

Le but de ce projet de semestre est de développer une plateforme pour traiter et analyser des images qui représentent des tissus pulmonaires potentiellement endommagés après un impact du réseau vasculaire. 

La société Deeplink est une startup d’intelligence artificielle spécialisée dans la technologie "chatbot". Cette entreprise utilise beaucoup le NLU (Natural Language Understanding) pour déterminer l’intention de l’utilisateur lorsqu’il communique avec le "chatbot". Actuellement la solution NLU est hébergée sur un serveur dédié et les échanges entre celle-ci et le client se font par requêtes http.

L’idée de ce projet consiste à faire tourner directement la solution NLU sur le navigateur du client. Avec une idée pareille, on assure qu’aucune requête contenant des données sensibles ne soit envoyée vers un serveur externe puisque l’inférence du modèle se fait sur le navigateur. En plus d’assurer une confidentialité bien plus sécurisée, si on déplace l’inférence du modèle on devrait obtenir de meilleure performance en termes de latence, d’extensibilité et on pourrait dans le futur développer des modèles personnaliser pour chaque personne.

The subject for this project is a dancer to which 6 motion tracking devices are firmly attached. The success of this project will demand the collection of the data from the sensors, and a successful recreation of the motion in a 3D space in Unity

The explicit main goals of the project could be stated as the following:

• Importing the data to unity in order to recreate the motions in a 3D space and saving the videos.
• Representing the sensors with different shapes.

As for the secondary goal, it was defined as the following :

• Using Unity to add customized behavior when the video is being made. Example: showing how one sensor is affecting another via a color code

The aim of this project is to develop a Web-based application to manage the different entities of the organization and their resources taking into consideration their dynamics and authorities. It will facilitate the saving and exchange of Human Resources data between three main users: 

- The General Admin 

- The Legal Representatives of the different companies 

- The Human Resources Managers 

Grâce à la "Gamification" de l’art, il est possible de toucher un plus grand public ainsi que de sensibiliser les plus jeunes aux éléments culturels.

Bobby, a therapeutic chatbot that offers music to improve the emotions and emotional states of individuals, was designed and implemented in this article. The preliminary results have shown a positive effect on the emotion strengthening, but no effect still on the emotion changing group.