Call for participation
Digital Creativity Centre
Universidade Catolica Portuguesa - School of Arts
3 – 28 July 2017
The Digital Creativity Centre (CCD), Universidade Catolica Portuguesa - School of Arts (Porto, Portugal) invites researchers from all over the world to join eNTERFACE'17, the 13th one-month Summer Workshop on Multimodal Interfaces. During this workshop, senior project leaders, researchers, and students gather in one single place to work in teams on pre-specified challenges for 4 weeks long. Each team has a project defined and will address specific challenges.
Senior researchers, PhD students, or undergraduate students interested in participating to the Workshop should send their application by emailing before 7th of May 2017(extended) to firstname.lastname@example.org (Visit Guidelines for researchers applying to a project).
Participants must procure their own travel and accommodation expenses. Information about the venue location and stay are provided on the eNTERFACE'17 website. Note that although no scholarships are available for PhD students, there are no application fees. The list of projects that participants can choose from is the following.
How to Catch A Werewolf, Exploring Multi-Party Game-Situated Human-Robot Interaction
Principal investigators: Catharine Oertel, KTH (PI), Samuel Mascarenhas, INESC-ID, Zofia Malisz, KTH, José Lopes, KTH, Joakim Gustafson, KTH
In this project we will focus on the implementation of the roles of the “villager” and the “werewolves” using the IrisTK dialogue framework and the robot head Furhat. To be more precise, the aim of this project is to use multi-modal cues in order to inform the theory of mind model to drive the robot’s decision making process. Theory of mind is a concept that is related to empathy and it refers to the cognitive ability of modeling and understanding that others have different beliefs and intentions than our own. In lay terms, it can be described as “to put oneself into another’s shoes” and it is a crucial skill to properly play a deception game like “Werewolf”.
KING’S SPEECH Foreign language: pronounce with style!
Principal investigators: Georgios Athanasopoulos*, Céline Lucas* and Benoit Macq*
Team Members: Guillaume Gustin, Alessandro Cierro and Robin Guerit. * ICTEAM-ELEN - Université Catholique de Louvain, Belgium
The principal investigators are developing the GRAAL1 project which is concerned with developing a set of tools to facilitate self-training on foreign language pronunciation, with the first target being learning French. The goal of KING’S SPEECH is to develop new interaction modalities and evaluate them in combination with existing functionality aiming to better personalize GRAAL to the taste and specificities of each learner. This personalization will rely on a machine learning approach and an experimental set-up to be developed during eNTERFACE’17. The eNTERFACE’17 developments could be based on a karaoke scenario where the song is replaced by some authentic sentences (extracts of news, films, publicities, etc.). Applications like SingStar (Sony) or JustSing (Ubisoft) could also serve as a source of inspiration, e.g., using a smartphone as a microphone while interacting with avatars.
The RAPID-MIX API: a toolkit for fostering innovation in the creative industries with Multimodal, Interactive and eXpressive (MIX) technology
Principal Investigators Francisco Bernardo, Michael Zbyszynski, Rebecca Fiebrink, Mick Grierson (EAVI – Embodied AudioVisual Interaction group, Goldsmiths University of London, Computing)
Team Members: Sebastian Mealla , Panos Papiotis (MTG/UPF – Music Technology Group, Universitat Pompeu Fabra), Carles Julia, Frederic Bevilacqua , Joseph Larralde (IRCAM – Institut de Recherche et Coordination Acoustique/Musique)
Members of the RAPID-MIX project are building a toolkit that includes a software API for interactive machine learning (IML),digital signal processing (DSP), sensor hardware, and cloud-based repositories for storing and visualizing audio, visual, and multimodal data. This API provides a comprehensive set of software components for rapid prototyping and integration of new sensor technologies into products, prototypes and performances.
We aim to investigate how developers employ and appropriate this toolkit so we can improve it based on their feedback. We intend to kickstart the online community around this toolkit with eNTERFACE participants as power users and core members, and to integrate their projects as demonstrators for the toolkit. Participants will explore and use the RAPID-MIX toolkit for their creative projects and learn workflows for using embodied interaction with sensors
End-to-End Listening Agent for Audio-Visual Emotional and Naturalistic Interactions
Principal Investigators: Kevin El Haddad (TCTS Lab - numediart institute - University of Mons, Belgium), Yelin Kim (Inspire Lab - University at Albany, State University of New York, USA), Hüseyin Çakmak (TCTS Lab - numediart institute - University of Mons, Belgium)
Team Members: Payton Lin (Research Center for Information Technology Innovation, Academia Sinica, Taiwan), Jaebok Kim (Human Media Interaction group, University of Twente, Enschede, Netherlands), Minha Lee (Human-Technology Interaction group - Technical University of
Eindhoven, Eindhoven, Netherlands), and Yong Zhao (6VUB-NPU Joint AVSP Research Lab - Vrije Universiteit Brussel,
Belgium & Northwestern Polytechnical University, China)
In this project, we aim at building a listening agent that would react with a naturalistic and human-like behavior and using nonverbal expressions to a user. The agent’s behavior will be modeled by and built on three main components: recognizing and synthesizing emotional and nonverbal expressions, and predicting the next expression to synthesize based on the currently recognized expressions. Its behavior will be rendered on a previously developed avatar which will also be improved during this workshop. At the end we should obtain functioning and efficient modules which ideally should work in real-time.
Cloud-based Toolbox for Computer Vision
Principal investigator: Dr. Sidi Ahmed MAHMOUDI from the Faculty of Engineering at the University of Mons. Belgium.
Team Members: Dr. Fabian LECRON, PhD, Faculty of Engineering at the University of Mons. Belgium, Mohammed Amin BELARBI, PhD Student, Faculty of Exact sciences and Mathematics, University of Mostaganem, Algeria, Mohammed EL ADOUI, PhD Student, Faculty of Engineering, University of Mons, Belgium, Abdelhamid DERRAR, Student in Master University of Lyon, France, Pr. Mohammed BENJELLOUN, PhD, Faculty of Engineering, University of Mons, Belgium, Pr. Said MAHMOUDI, PhD, Faculty of Engineering, University of Mons, Belgium.
Nowadays, images and videos have been present everywhere, they can come directly from camera, mobile devices or from other peoples that share their images and videos. The latter are used to present and illustrate different objects in a large number of situations (public areas, airports, hospitals, football games, etc.). This makes from image and video processing algorithms a very important tool used for various domains related to computer vision such as video surveillance, human behavior understanding, medical imaging and database (images and videos) indexation methods. The goal of this project is develop an extension of our cloud platform (MOVACP) developed in the previous edition of eNTERFACE’16 workshop. The latter integrated several image and video processing applications. The users of this platform can use these methods without having to download, install and configure the corresponding software. Each user can select the required application, load its data and retrieve results, with an environment similar to desktop. Within eNTERFAC’17 workshop, we would like to improve and develop four main tools for our platform:
- Integration of the major image and video processing algorithms that could be used by guests to perform their own applications.
- Integration of machine learning methods (used for images and videos indexation) that exploit the uploaded data of users (is they accept of course) in order to improve the results precision.
- Fast treatment of data acquired from IOT systems.
- Development of an online 3D viewer that could be used for the visualization of 3D reconstructed medical images.
- Fast treatment of data acquired from distant IoT systems.
Keywords cloud computing, image and video processing, video surveillance, medical imaging.
Across the virtual bridge
Thierry RAVET (software design, motion signal processing, machine learning), Fabien GRISARD (software design, human-computer interface), Ambroise MOREAU (computer vision, software design), Pierre-Henri DE DEKEN (software design, game engine) - Numediart Institute, University of Mons, Belgium.
Team members :
Mickael TITS (Motion Signal Processing, Machine learning), François ROCCA (computer vision, behaviour analysis), Radhwan BEN MADHKOUR (computer vision, software design), - Numediart Institute, University of Mons, Belgium, François ZAJEGA (3D design, Game engine) - ARTS2, Mons, Belgium.
The goal of the project is to explore different ways of creating interactions between people evolving in the real world (local players) and people evolving in a virtual representation of the same world (remote players). This latter one will be explored thanks to a virtual reality headset while local players will be geo-located through an app on a mobile device. Actions executed by remote players will be perceived by local players in the form of a sound or visual content and actions performed by local players will impact the virtual world as well. Local players and remote players will be able to exchange information with each other.
Keywords: Virtual world, mixed reality, computer-mediated communication .
ePHoRt project: A telerehabilitation system for reeducation after hip replacement surgery
Principal investigators: Yves Rybarczyk (Nova University of Lisbon, Portugal), Arián Aladro (Universidad de las Américas, Ecuador), Mario Gonzalez (Health and Sport Science from University of Zaragoza - Spain), Santiago Villarreal (Universidad de las Américas - Quito, Ecuador), Jan Kleine Detersa (University of Twente in Human Media Interaction)
This project aims to develop a web-based system for the remote monitoring of rehabilitation exercises in patients after hip replacement surgery. The tool intends to facilitate and enhance the motor recovery, due to the fact that the patients will be able to perform the therapeutic movements at home and at any time. As in any case of rehabilitation program, the time required to recover is significantly diminished when the individual has the opportunity to practice the exercises regularly and frequently. However, the condition of such patients prohibits transportations to and from medical centres and many of them cannot afford a private physiotherapist. Thus, low-cost technologies will be used to develop the platform, with the aim to democratize its access. For instance, the motion capture system will be based on the Kinect camera that provides a good compromise between accuracy and price. The project will be divided into four main stages. First, the architecture of the web-based system will be designed. Three different user interfaces will be necessary: (i) one to record quantitative and qualitative data from the patient, (ii) another for the therapist consulting the patient’s performance and adapting the exercises accordingly, and (iii) for the physician having a medical supervision of the recovery process. Second, it will be essential to develop a module that performs an automatic assessment and validation of the rehabilitation activities, in order to provide a real-time feedback to the patient regarding the correctness of the executed movements. Third, we also intend to make use of a serious game and affective computing approaches, with the intention of motivating the user to perform the exercises for a sustainable period of time. Finally, an ergonomic study will be carried out, in order to evaluate the usability of the system.
Big Brother can you find, classify, detect and track us ?
Principal investigators: Marc Décombas, Jean Benoit Delbrouck (TCTS Lab - University of Mons, Belgium)
In this project, we will build a system that can detect, recognize objects or humans and describe them as much as possible on video. Objects may be moving as well as the people coming in and out of the visual field of the camera(s). Our project will be split into three main tasks :
- detection and tracking
- people re-identification
- image/video captioning
The system should work in real time and should be able to detect people and follow them, re-identify them when they come back in the field and give a textual description of what each people is doing.
Networked Creative Coding Environments
Principal investigators: Andrew Blanton, Digital Media Art at San Jose State University in San Jose California
As a part of ongoing research Andrew Blanton will present a workshop using Amazon Web Servers for the creation of networked art. The workshop will demonstrate sending data from Max/MSP to a Unix based Amazon Web Server and receiving data into a p5.js via websockets. The workshop will explore the critical discourse surrounding data as a borderless medium and the ideas and potentials of using a medium that can have global reach .
Study of the reality level of VR simulations
Principal investigators: Andre Perrotta, UCP/CITAR
Team members : Pedro Pestana (UCP/CITAR), Jorge Cardoso (UCOIMBRA/CITAR), Celio Jonas Monteiro PhD student - UCP/CITAR, Armando Ramos PhD student - UCP/CITAR, Ricardo Ferreira - PhD student - UCP/CITAR, Bernardo Liborio
Msc student - UCP/CITAR
We propose to develop a VR simulation based on 360o video, spatialized audio and force feedback using fans and motors, of near collision experiences of large vehicles on a frst person perspective, to be experienced by users wearing head-mounted stereoscopic VR gear in a MOCAP (motion capture) enabled environment that enables a one-to-one relationship between real and virtual worlds.
AUDIOVISUALY EXPERIENCE THROUGH IMAGE HOLOGRAPHY [canceled]
Principal investigators: Maria Isabel Azevedo ( ID+ Research Institute for Design, Media and Culture,University of Aveiro), Elizabeth Sandford-Richardson ( University of the Arts, Central Saint Martins College of Art and Design), Francisco Carneiro Pacheco ( LivingSocial UK), Nuno Janardo ( PhD Candidate University of Lisbon, Institute of Art History), Martim Nunes Pinto ( Catholic University of Portugal in Porto, the School of Arts.)
Today in interactive art, there are not only representations that speak of the body but actions and behaviours that involve the body. In digital holography, the image appears and disappears from the observer's vision field; because the holographic image is light, we can see multidimensional spaces, shapes and colours existing on the same time, presence and absence of the image on the holographic plate. And the image can be flowing in front of the plate that sometimes people try touching it with his hands.
That means, to the viewer will be interactive events, with no beginning or end that can be perceived in any direction, forward or backward, depending on the relative position and the time the viewer spends in front of the hologram.
In this workshop we are proposing an audiovisual interactive installation composed by four digital holograms and spatial soundscape. When viewers move in front of each hologram, different sources of sound are trigger. The outcome will be presented in the last week of July with an invited performer. We are looking for sound designers and interaction programmers.
Keywords: Digital holographic image, holographic performance, sound spatialization, motion capture
Principal investigators : Ivan Franco (IDMIL / McGill University)
Prynth is a technical framework for building self-contained programmable synthesizers, developed by Ivan Franco at the Input Devices and Music Interaction Lab (IDMIL) of McGill University. The goal of this new framework is to support the rapid development of a new breed of digital synthesizers and their respective interaction models.