In this paper we show how an artificial intelligence system is able to recognize actions of a group of people during remote meetings. It is based on the combination of several basic detections. We explain why it is a very important task in order to help a coach manage teams. In particular, we explain how the recognition of discussions allows to measure the collaboration between participants or possibly their isolation. We detail the action of Eye Contact where two people look at each other during a discussion. We present the algorithm of the module of our system in charge of the recognition of an Eye Contact action and the difficulties raised by this module. We conclude with the heuristics rules that would allow to detect higher level of group actions from individual actions in brainstorming sessions.

This paper investigates the use of digital collaborative tools and live streaming systems to support student initiation into scientific research. A Computer-Supported Cooperative Work in Design (CSCWD) system was implemented to aid two key phases: collaborative planning in a digital workspace and experimental execution in laboratories. In the first phase, multitouch surfaces enabled students to collaboratively define research problems and protocols, fostering idea generation and consensus building. The second phase utilized video capture and streaming systems for documentation and remote participation in experiments. While these tools showed potential, challenges such as camera resolution, latency, and feedback limitations emerged. Initial findings suggest these technologies support research skill development, but further improvements are needed to enhance usability and engagement for both local and remote participants. Future work will focus on addressing these technical and interaction challenges to better support hybrid learning environments.

This paper presents the concept of Digital Ji-in, a framework for managing permanent digital identifiers whose structure is based on a peer to peer network. The origin of the term Digital-Ji-in is explained in the paper. We describe several types of identifiers and explain why their management is already an important issue. We show that some identifiers have been reused, creating significant problems. We propose a framework for maintaining identifier information safely over a long period of time. We present its structure based on a secure network of peers. We also describe some use cases of this framework. However, as only identifiers are recorded mechanically, problems may arise if data should be associated with them. We discuss this issue at the end of the paper.

In this paper, we describe an AI-based system that recognizes the activity status of several people from video streams during brainstorming meetings. Deep learning is often used to recognize video characteristics but requires a huge amount of computer resources. This makes it difficult to keep track of the activities of multiple people whose circumstances change. On the other hand, many trained models of one person’s motion recognition have been developed and are available. We propose to use the existing technology but to be able to do that we need to identify a single person’s activities within a group context. This is achieved by segmenting the video and cropping the area with a person, identifying the activity using pre-existing trained models. The activity of the group is recognized by a production rule system based on individual activities. To achieve our goal, we introduce the concept of atomic action to describe activities and propose categories of atomic actions. High-level collaborative categories that indicate the status of a group during collaborative meetings are based on the CIAO model. This paper ends with the results of the first experiments we conducted using video recordings of actual students’ work sessions.

In this paper, we describe the design and the management of an agent-based system that supports distributed brainstorming activities. The support system is a highly coordinated IoT application composed of many locally installed interface devices, multimedia communication functions, and cloud functions that process application logic and store meeting data. The system is designed to support a variety of brainstorming sessions, so its functionalities must be modifiable and enable the system to be adapted to different environments and user requirements without any loss of performance. System accessibility should be also ensured from any location for any user. These constraints require a flexible and usable support system. We further discuss the aspects of flexibility and usability that are important in a support system for distributed brainstorming, from which we propose a conceptual schema for flexible and usable support systems. To realize this schema, we present a resource-oriented architecture that can modify the brainstorming support system’s structure and functions. Flexibility is achieved thanks to an agent-based system that manages resources and operates on them according to users’ requests. We also describe the system architecture, which is organized around a set of channels dedicated to different services proposed to the users. We present in detail a video channel that ensures user awareness during synchronized activities. We then conduct several experiments verifying the usability of important channels in the architecture and present the results of these experiments. Finally, we discuss experimental scenarios that show how the system owes its adaptability to management based on an agent organization that supports distributed brainstorming and other activities.

In order to get better results during brainstorming activities participants must respect some rules when they write notes about ideas and when they consider notes written by somebody else. We argue that the respect of these rules can be verified by a multi-agent system analyzing videos and notes produced by the participants in real time. This system can simplify the role of the meeting facilitator. Feedback is sent individually or is addressed globally to the entire team. This paper presents considerations about the necessity of rules, the structure of a multi-agent system for analyzing the respect of those rules and how experiments could show the level of acceptance by people of such a system. The goal of this prospective research is to add a non intrusive system during brainstorming sessions in order to enhance the quality of results.

A remote brainstorming support system is a highly coordinated IoT application composed of many locally installed interface devices, multimedia communication functions, and cloud functions that store meeting data. In this paper, we describe an agent-based remote brainstorming support system for exchanging multimedia information among distributed teams according to user requests. We aim to (1) develop multimedia channels to communicate brainstorming results obtained from and presented on various devices among distributed teams, (2) establish a novel mechanism to add newly developed channels to the system on demand during a meeting, and (3) scale up the performance of remote brainstorming support systems.

The ubiquitous technology attracts attention widely to support people life effectively. However, developing such application systems impose heavy burden to developers because they have to combine very wide fields of technologies such as sensor devices, mechatronics, pattern recognition, machine learning, activity understanding, decision making, etc. In this paper, first, we propose a method to connect device, data and programs to an intelligent agent platform. Next, we describe a ubiquitous system prototype that supports meeting sessions. We have developed this prototype as a test bed. It is is still a small example for testing a ubiquitous agent system, but it has potential ability to support meetings because it favors collaboration between people and ubiquitous agents.

The motivation of our research is the design and the implementation of tools supporting the activities of several distant teams that have to collaborate for defining the specifications of a project. The paper presents the design of synchronous brainstorming activities between two distant teams mediated by two channels: a TV channel and a resource channel. Via the TV channel, cameras film a team during a session and computer applications allow to record and transmit video sequences to the distant team. During sessions, each team interacts with a large multi-touch device and manipulate virtual post-it notes displayed on both surfaces. Via the resource channel, displays each device are synchronized. The paper also presents the architecture of the main system involving several tactile devices, individual devices like tablet pcs and collaborative ones like interactive tables or whiteboards. The experiment conducted between two team situated in Japan and in France and conclusions about this experiment are described in the last sections.

The paper discusses the open multi-agent architecture of a system comprising a large multitouch multi-user graphics table, a large interactive board, and a number of peripheral devices including tablets, smartphones, or earphones, used for supporting collocated collaborative work. The system comprises several computers connected in a network, or through WiFi or Bluetooth interfaces. The system software is organized around two multi-agent platforms. The first one, JADE, written in Java is used to integrate agents directly into the toolkits which use Java. The second one, OMAS, written in Lisp is used to implement conversational intelligent agents. The paper discusses the role and the use of both platforms and how they are connected to work together. Although developed in a project addressing collaborative Preliminary Design, the system and its architecture can be used in any application composed of sequences of phases.

International audience

In recent years, there has been an increased interest for research on computer-supported cooperative work performed in collaborative interactive spaces. The TATIN-PIC project envisions a true multi-surface collaborative work environment with an interactive tabletop, an interactive board display, tablet PCs, and smartphones. In this paper, we first present the middleware based on a multi-agent architecture which use in our implementation, and then we detail how voice-controlled personal assistant agents can be implemented to provide unique interactions within a multi-surface environment.

The recently launched TATIN-PIC Project at the University of Technology of Compiègne is a multi-modal collaborative work environment for teams performing preliminary design. In this paper, we present the design and reasoning for the plans of our groupware system. Effective preliminary design teamwork requires alternating stages of discussion, exploration, and diverging ideas, coupled with stages of focus, decision-making, and convergence. We believe providing a design team with an interactive tabletop can fulfill the first role and an interactive whiteboard can satisfy the second. Users will be able to switch seamlessly between the two, shifting content from one to the other when necessary. In this paper, we begin with a discussion of the advantages and characteristics of preliminary design. We continue with a review of the state-of-the-art in interactive, board-sized surfaces. We discuss experiments and results from TATIN, the precursory project to TATIN-PIC. Finally, we introduce our plans for an innovative work environment. Its multi-agent software architecture will be capable of supporting voice commands, tablets, smartphones, and personalized assistants capable of ontological reasoning about the project work.

In this paper, we first discuss the necessity of a support system for distributed multi-lingual development teams, to bridge the gaps between different languages, cultures and social systems of a client and a vendor when defining a requirement specification of an application system. Next, we propose the design of an agent-based support system for defining requirement specifications, consisting of collaborating design assistant agents, common design support functions and a language translation web service. To implement the approach we plan to use OMAS, a platform for developing multi-agent systems. OMAS has features for describing multi-lingual ontologies and multi-lingual representations of dialogs. It also provides a platform for testing multi-cultural cooperation over the Internet. Some experimental applications are being developed currently using different natural languages. In this paper, we propose an agent-based support system for requirement specification definition as one such application. The proposed system is a web-based and agent-based application and supports a multi-lingual software design team using a network of personal design assistant agents.

This paper discusses harmonization issues between the management of resources stored locally by users in personal repositories and the management of resources shared by a community which they belong to. The solution consists in maintaining the same kind of indexing. We discuss some requirements that the motivation implies and we present some choices that have been done for engineering the system that allows to manage different types of resources and that allows to describe them with ontological elements. This paper focuses on a knowledge-based representation of resources that supports the service that the community can bring to its members. We develop our approach of semantic indexing. Some constraints due to the nature of a distributed index are also presented. Peer to peer based communities rely on a decentralized system. The paper also presents the system engineering of a prototype that gives access to the resources. It explains how the architecture is structured and the role of its layers. The Web design of the front-end guarantees an anywhere and anytime access to the resources. The system supports an easy scalability thanks to its underlying architecture based on web services.

In this paper, we first present an analysis of a multi-language application from a cross-cultural point of view. Then we analyze the specific activities of a multilingual collaborative design and the automatic translation system that it requires. We also propose an architecture of a design support system producing the UML design of applications and allowing a multilingual collaboration. We present the first results of an experiment with such a system.

The increasing number of available resources that may be used during e-learning scenarios can show problems of access, management and sharing of these resources. An e-learning application deals with the same problem of relevance to the Web. The semantic Web recommends standards for providing a solution to these issues. One main feature of this new Web generation is the use of semantics based on ontologies. This paper presents an approach for indexing learning resources by a specific LOM ontology. We also present the tool that we have built allowing to describe learning objects and helping to search for them.

This paper describes an approach for a composition of Web services based on their semantic descriptions. The process section of OWL-S service descriptions is built with references to ontology concepts which represent service input and output data types. We present an engine that receives a request containing a concept (OC) corresponding to a service output and a set of concepts (ICs) corresponding to a service inputs. The engine produces a sequence of services whose first element has ICs as inputs and whose last element has OC as output. The result of the composition is described as a BPEL process.

As each university owns several web sites dedicated to face-to-face lectures support or distance learning training, courses can no more be considered as global presentations of theory elements. Organizations involving different actors, contribute to produce documents and information and doing so must share knowledge. We present in this paper a way to index documents inside a memory adapted to e-learning situations. This indexing mainly requires concepts representing the pedagogical notions of the domain of training. It requires also some concepts that are typical of learning activities.

We present work done on the use of multiple intelligent agents as teaching agents, and discuss some more general applications for agent technology in education.