Everyone who wears glasses knows: finding the best pair for your face can be a daunting task… The VESPA project, led by researchers at IMT Lille Douai and hosted by the Teralab platform, is working with Cape Trylive to develop machine learning algorithms that can determine users’ facial characteristics during virtual online fittings and recommend products based on their morphology.
Do you have a round face? Then the best glasses for you would be angular and narrow designs. Do you have a square face? Then try butterfly-shaped glasses! Based on your face shape, certain types of frames will look better on you than others. For a few years now, plug-ins on opticians’ websites have allowed users to try on glasses virtually. However, they do not offer features to help users find the best frames for their face.
The VESPA project, led by Jacques Boonaert and Stéphane Lecoeuche at the IMT Lille Douai Computer and Control Sciences Research Unit, in collaboration with Pascal Mobuchon of Acep Trylive, a virtual fitting plug-in supplier, aims to develop a machine learning algorithm capable of identifying users’ face shapes and recommending glasses based on their morphological characteristics. “Acep Trylive has developed technology for establishing the key points on users’ faces while they are filmed via webcam during virtual fittings,” Stéphane Lecoeuche explains. “The goal of the VESPA project is to develop algorithms capable of analyzing the position of key elements on faces to determine the user’s morphology.” All the data used in the context of this project are anonymized and hosted on the secure, neutral and sovereign Teralab platform. TeraLab also provides researchers with tools for processing these data sets.
The algorithms developed by the researchers follow a supervised learning logic. “This means that we submit a set of images labeled by human experts to the algorithms, which determine if the person’s face is round, square, oval…” Jacques Boonaert explains.
In addition, the key points that Acep Trylive’s software automatically establishes on the user’s face provide the algorithm with a set of descriptors. For example, a descriptor could be a measurement of the key points on the face: this could be the height of the forehead, the shape of the chin, the width of the face or the jaw, etc. Based on the data labelled by human experts, the algorithm will determine which descriptors are most relevant in recognizing the user’s morphology. “We have tested sub-sets of descriptors. In all, there are over 20,” says Stéphane Lecoeuche. “The algorithms then ascertain the influence of the descriptors on their own to define the best morphological characterization.”
There are three of these morphological classification algorithms. One focuses on the shape of the jaw, the second on the face shape, and the third on the width of the forehead. Digital descriptors of the user’s hair, eye and skin color are also used to propose the most suitable colors for the frames. All of this data is then merged to create recommendations for glasses adapted to the user based on these characteristics.
The Acep Trylive software highlights the key points on users’ faces (in blue)
Recommending products based on users’ morphology
Based on the monitoring of behavior and consumers’ history, the researchers were able to determine which glasses each internet user preferred: “Someone tries on a first pair of glasses, then a second, comes back to the first, tries a third pair, then again comes back to the first… In observing this sequence, we can determine which product the person prefers!” Jacques Boonaert explains.
The learning algorithm takes this fitting history into account and statistically consolidates the glasses the user preferred by morphology type. “Thanks to the data from thousands or even hundreds of thousands of fitting sessions, the algorithm makes the connection between the face shape and the products that were tried on,” says Stéphane Lecoeuche. Therefore, for each new user who tries the application, the morphological analysis algorithms will determine their face shape and, depending on the choices of users with a similar morphology, and the recommendation engine will propose a set of products most likely to please them.
Seeking a partner company for further development
“We also had to work on the product’s geometric classification. The problem is that we have not been able to access the data from opticians’ catalogs, which classify glasses by shape, style, color…” Stéphane Lecoeuche explains. The researchers had planned to work on the association between the customer’s morphological characteristics and the products’ geometrical characteristics. While the algorithmic analysis of the facial morphology provided good results, the lack of data on the frames has limited their objectives.
“The other difficulty is that we do not have access to users’ final purchase decisions,” Jacques Boonaert adds. “We are aware that this is sensitive data for businesses. This is why we would like to create a strong partnership with an optician in order to further develop this project.” The researchers wish to implement a second phase of experimentation with a partner company, in which the algorithm could integrate the users’ buying decision and the products’ geometric classification. In the meantime, while waiting to find a company that would like to integrate the project consortium, the research team is continuing its work in machine learning geared towards industry and trade.
https://imtech-test.imt.fr/wp-content/uploads/2018/01/scott-van-daalen-202643.jpg27304096I'MTechI'MTech2018-03-30 17:20:542018-03-30 17:31:43Facial morphology analysis algorithms for choosing glasses
[dropcap]T[/dropcap]he climate is warming and changing: we must act. On the French version of the website The Conversation, climate change specialist Christian de Perthuis recently applauded the introduction of the carbon tax in France in 2014 and its gradual increase.
Referring to the “Quinet value” (around 90 euros), the author suggests that reaching this price would “guarantee” that greenhouse gas emissions would decrease four-fold. When they accumulate in the atmosphere, these emissions disrupt the climate’s balance.
France therefore would appear to be on the right path, seemingly part of the “narrow circle” of model students. Sweden, with its carbon tax of 110 euros, is cited as an example. Inequalities in terms of access to energy are offset by the energy voucher benefit. There’s only one hitch: the carbon tax is very unpopular.
Dependence on fossil fuels
This is an appealing example, but in fact it is deceiving.
First, we must remember that Sweden depends less on fossil fuels than France, representing approximately 30% of its primary energy balance against 50% in France. And France is much more dependent than the figures show.
We must keep in mind that “primary energy” refers to the energy produced within a country, to which we must add imported energy. This amount is different from the “final energy”, which refers to the energy charged to the consumer. Between these two categories is where the losses are hidden. However, there is a 70% difference between primary and final energy for nuclear energy, which is crucial in the French energy mix.
In the light of these figures, we understand why it is easier for Sweden to combat these greenhouse gas emissions.
Also, the transition Sweden has achieved over the past decades did not only rely solely on the carbon tax. While it is true that the country went from 70% of fossils in the 1970s to approximately 30% in its primary energy balance, this was due to a set of combined measures, including grants and the development of local channels; but also due to a serious economic crisis in the 1990s.
This led to devaluation, which resulted in a considerable increase in the price of fossil fuels, which brought about significant and lasting changes in the local market conditions. Finally, Sweden developed electricity production primarily based on hydropower and nuclear energy. And the “decarbonization” was also brought about by the strong development of district heating networks and significant incentive policies for renewable energies.
The tax is must therefore be understood as part of a clear and comprehensive strategy, based on strong political support, which is not the case in France. On June 10, 2016, the five main political parties in Sweden concluded an agreement on the country’s energy policy.
One of the specific objectives was that Sweden would no longer emit greenhouse gases by 2045, and that the national electrical production would rely entirely on renewable energies as early as 2040. Measured in primary energy, Sweden already relies on 36% renewable energy: this is much more than France, which has stagnated in this area for the past 30 years (13 Mtoe in 1970 on a balance of 170 Mtoe, 24 on a balance of 260 Mtoe in 2016).
There is nothing to show that the tax played the role attributed to it by Christian de Perthuis, quite the contrary. We can also note that energy policy is a topic that is regularly absent from French public debate.
Outsourced emissions
Yet in Sweden, things are not as rosy (or green!) as they seem. A significant part of the country’s economy is now devoted to the tertiary (services) sector, and the share of the industrial sector has declined, as in France. Services account for 72% of GDP and 80% of the labor force. These figures are comparable to those in France (80% and 76% respectively).
This means that Stockholm is increasingly buying the goods consumed on its territory, like its electronic products, elsewhere. Therefore, Sweden outsources the greenhouse gas emissions related to the manufacturing of these products. In 2008, WWF estimated that 17% of emissions should be added to Sweden’s balance to obtain the net total of its emissions. This is still better than France, which must add 40%.
For all its talk, France is not a leader in climate issues.
Paris and much of French media prefer to criticize German coal for its harmful emissions, despite Germany having reduced these emissions by 25% since 1990. While many commentators are focusing on Germany’s failures (perhaps to better excuse our own failures?), we must remember that German emissions have dropped from 1041 MtCO₂ in 2000 to 901 in 2015.
Germany, with its less service-based economy, draws its fuels from its own soil; therefore, it does not outsource its emissions. Although German emissions stopped declining in 2009, their decarbonization policy remains unchanged; it has simply introduced the additional constraint of phasing out nuclear power. And it is doubtful that Germany will succeed in doing what also appears impossible in France (and elsewhere): reconciling infinite growth with respect for the planet.
French companies have bad report cards
All the countries in the world cannot continue to pass their emissions on to their neighbors. Therefore, the decoupling of economic growth from greenhouse gas emissions mentioned by Christian de Perthuis in his article certainly appears to be an illusion.
In reality, France is the 7th largest global contributor to climate change. This is explained by the mass of cumulative emissions over a long period of time, keeping in mind that what causes global warming is not the annual emissions of greenhouse gas, but rather their gradual accumulation over the past 150 years.
The warming is a function of the concentration of greenhouse gases in the atmosphere, and thus of the overall carbon stock there. Although France now emits relatively little in comparison, it has emitted large amounts in the past. We would expect France to set an example, as it committed to do under the Kyoto Protocol.
We must emphasize that the policies of French multinationals are very far from respecting the recent commitments of the Paris Agreement to combat climate change: these policies lead to a rise of 5.5 °C. Finally, Sweden and France are taking risks in committing to nuclear energy, despite the statistical certainty of an accident revealed by the revised calculation methods following the Fukushima disaster.
A political issue
Distributing a voucher for 150 or 200 euros to help the poorest individuals cope with their dependency on fossil fuels is indecent in a context of increased wealth among the wealthiest individuals and the stagnation of low wages in developed countries. The most affluent will continue to consume more and more, thus emitting more and more greenhouse gases.
Finally, there is the question of how Christian de Perthuis’ article can define a “good policy” based solely on economic analysis, without including a multi-disciplinary perspective and a public consultation process.
The wealthiest individuals and high-value added activities, which “drive” growth, will pay the tax without even noticing, and will continue to feed the machine, consuming more and more; while those with the lowest incomes, or those for whom fuel represents a significant portion of their budgets, will be greatly impacted. None of this makes any political sense.
In budgets, the priority continues to be given to roads and cars (+16% in 1990-2014 compared to -14% for rail tracks). And yet we want to penalize those who use them? Do we really want to save the climate?
The right approach lies in popular mobilization and creative energies; this is a political problem, not only an economic one. By ignoring this, we avoid putting the real obstacles on the table, such as market balances between major operators (who want to keep their shares, arguing that they need time to change) and innovators that are struggling to make a name for themselves.
In November 2017, IMT became an associate partner of EIT Health, a European program for improving research on new technologies in e-health. Bernadette Dorizzi, director of research and the doctoral program at Télécom SudParis, explains the objectives of this partnership, and how it will contribute to developing IMT projects on e-health topics.
What is EIT health and what are its objectives?
Bernadette Dorizzi – EIT Health is a consortium of 50 core partners and 90 associated partners from 14 different EU countries. It brings together companies such as Air Liquide, Sanofi-Aventis, Siemens Healthineer, research organizations including CEA and INRIA and universities like Imperial College in London, the University of Copenhagen and the Technical University of Munich. EIT Health is arranged in national “nodes” with European governance. The French node is extremely active and won the most projects last year!
This consortium offers funding and connections within an ecosystem of startups, manufacturers and academics to develop and enhance projects in the area of e-health. For example, a researcher who is developing a concept can work with a startup to develop a prototype, and then with a larger company that will enhance and distribute the device. Other projects, of larger financial amounts, are aimed at addressing societal problems, such as the autonomy of dependent persons at their homes. They are carried out by large companies in association with SMES, whose participation is highly valued. The overall objective is to take research out of the laboratories, so it can make a more significant impact on society.
In addition to events and meetings for sharing ideas and projects, EIT Health proposes training sessions on major issues related to innovation and entrepreneurship. For example, Sanofi is currently developing a training program on the GDPR, the General Data Protection Regulation, which will come into effect in Europe as early as April of next year. It is an interesting symbiotic relationship: the academics are not the only ones receiving training, manufacturers are too!
Our goal within EIT Health is of course to promote our research projects. At IMT, we have a cross-cutting program on “Health, autonomy and well-being”, and we are looking for partners for developing projects.
What areas is EIT Health involved in?
B.D. – EIT Health is involved in new technologies and big data in the e-health sector, and particularly issues such as preventive care, home support for the elderly, improving the employability and autonomy of dependent persons, and the care provided to patients with chronic diseases.
A new call for proposals focuses on “wild card” projects, which are very innovative and high risk, on very precise subjects. In 2018, they focus on resistance to antibiotics and the use of data in personalized medicine.
What added benefit does IMT bring to these issues?
B.D. – IMT is offering EIT Health its ecosystem, which represents a wealth of research and innovation in health technologies and services, in a field where it is a national player and is internationally recognized. For example, a number of our researchers are working on connected objects in health, for the quality of life and independence of vulnerable individuals. We are currently conducting a project called Solsens, financed by the “Health, autonomy and well-being” seed fund in 2017. This inter-school project looks at flooring technology that could detect walking movements and falls and could send this information to a smartphone or computer. These products are manufactured by the German company Future Shape. Our researchers have worked on this concept to develop serious games that reproduce the walking path that was taken and want to develop new applications for this connected flooring. They would like to find a partnership with a manufacturer to develop this device on a larger scale.
The spin-off companies and incubated companies at the different IMT campuses also represent a significant contribution to the EIT Health dynamic. As part of a project on anonymizing personal health data, a team of researchers from IMT Atlantique created a small startup and are seeking to further establish themselves in order to offer their services to e-health stakeholders. This is exactly the type of situation in which our collaboration with EIT Health could be beneficial.
In addition, various technological platforms, such as Teralab, a big data and artificial intelligence platform featuring an authorized system hosting health data, clearly illustrate the added benefit IMT brings to EIT’s activities.
For the time being, IMT is an associate partner. If the partnership goes well, and we obtain good results from the current projects, we will consider becoming a core partner in order to carry out projects on a larger scale.
https://imtech-test.imt.fr/wp-content/uploads/2018/03/illu-EIT-Health-noir-et-blanc.jpg27263425I'MTechI'MTech2018-03-27 15:55:322018-04-12 15:29:19IMT becomes an associated partner of EIT Health
[dropcap]D[/dropcap]igital social innovations (DSI) are booming in Europe, empowering people to solve problems in areas as diverse as social inclusion, health, democracy, education, migration and sustainability. Examples include civic tech, neighbourhood-regeneration platforms, collaborative map-making, civic crowdfunding, peer-to-peer education and online time banks. A wide range of organisations support DSIs, through offering consultancy services, network access, funding, resources and skills.
The UK-based NESTA is one of the central think tanks in the field, as well as the coordinator of the EU-funded project DSI4EU. At the European level, different schemes exist to support social innovations and also DSIs, such as the Social Innovation Competition, whose 6th round took place in Paris on March 20 this year. Many events, festivals and conferences are also being organised, such as the Social Good Week in Paris or the Ouishare Fest, which was born in France and is now an international event.
While significant time, effort, and resources are spent on these activities, there are some obstacles to their development and efficacy in tackling the big challenges of our times, which seem necessary to address.
1. Questioning openness
Many DSIs emphasise participation and transparency, but the use of open-source software remains limited, at least in France. The openness of a platform is an important indicators about its capacity to encourage participation, by decentralising power, enabling others to access, replicate, and build upon the source code. Proprietary software, on the other hand, raises questions about the extent to which it is being manipulated by the innovator. Valentin Chaput, the editor of the site Open Source Politics states: “When we do not master its code, it is the authors of this code who control us”.
2. What happens to user data?
Social entrepreneurs often struggle to build sustainable business models that will ensure their autonomy and independence. There exist different business models through which DSIs generate income. One of these is the commercialisation of user data. Here, the main problem is not commercialisation per se (although to prevent it would be preferred), but how the background business model is communicated with the users.
To have information, users need to read in detail the platform’s “terms of use”, which are often not communicated by an attractive design. As a consequence, users can easily skip this part, due to ignorance or lack of interest. Platforms should be more transparent about their business models, and communicate these with the audience in a user-friendly way. This will also reduce some users’ hesitations in involvement, caused by a lack of trust.
3. Systemic change or short-term relief?
There is also a deeper concern about the sharing economy. Evgeny Morozov, author of Net Delusion: The Dark Side of Internet Freedom, wrote, “it’s like handing everybody earplugs to deal with intolerable street noise instead of doing something about the noise itself”. Sometimes this is also valid for DSI. How can innovations that can bring a systemic change be distinguished from system-enhancing ones? It is not meaningful to categorise platforms as systemic ones and others, as there are different shades of grey between purely black or white.
But there is some scope for thinking deeper, by observing the activities of platforms. For example, Humaid is a crowdfunding platform in which people with disabilities or their caregivers can raise money to purchase necessary assistive technologies. In so doing, Humaid in a sense reproduces exclusionary practices in the society by taking people with disabilities as objects of charity, rather than as individuals with rights and freedoms, as outlined in the UN Convention on the Rights of Persons with Disabilities.
Another example is from the sharing economy. Karos, a car-sharing platform launched a year ago, provides the option of “ladies only” car sharing. In doing so, doesn’t Karos reproduce existing practices that give rise to inequalities in the first place? Rather than using information and communication technologies (ICTs) to alleviate inequalities embedded in societies, such initiatives enhance existing norms and exclusionary barriers. Addressing big challenges require awareness raising and educational activities around rights and freedoms.
4. The struggle of traditional civil-society organisations
Established civil-society organisations that have field-specific experience with targeted populations, and who are involved in social movements and awareness raising activities can have an important role in systemic change, but most of them find themselves in a vulnerable position faced with digital platforms. For example, some are facing competition from start-ups that build resources and finances from the digital sector. Digital competences of the new economy and traditional associations’ field-specific experiences should find spaces of synergy building. But there are barriers to the successful building-up of such spaces, sometimes due to polarised ideological worlds between non-profits and organisations of the digital economy.
5. Under-engagement of users
There is also the important issue of attracting users to these platforms. Most DSI platforms rely on civic engagement, which could be for volunteering, providing skills, information, services, goods, opinions. At the same time, the online world is likely to reflect the economic, social and cultural relationships in the offline world – a research paper by Alexander Van Deursen and Jan Van Dijk of the University of Twente sheds light on this question.
This suggests that the DSI users could be those who are already active in civic life in the offline world, as indicated by the research of Marta Cantijoch, Silvia Galandini, and Rachel Gibson. If this is the case, DSIs can strengthen existing divides instead of alleviating them. To be able to develop effective and informed policies, more research about the nature of users, their engagement patterns in different platforms are needed, but there are obstacles on the way; most important is the lack of data.
6. Lack of data in a world of ‘big data’
The lack of data on users and the ecosystem are serious barriers to carry out research on DSIs and their potential to address big challenges. Platforms do not share data due to privacy and confidentiality reasons. Or, as in the case of France, regulations about data collection can prevent research about the users of DSI. At the national and EU levels, initiatives to collect and standardise data are much needed, so that researchers can have access to essential data about the use of and participation in DSI. This is also important to carry out research on the specific capabilities of different EU countries on DSI and develop means to transfer good practices and make use of potential synergies.
7. Fascination with (rapid) impact measurement
For investors, funders, and social entrepreneurs, social impact measurement is essential. But this can be problematic, complex and difficult issue. What’s more, it is important to remember a quote from William Bruce Cameron: “Not everything that can be counted counts. Not everything that counts can be counted”. In addition, sometimes time pressures result in employing vague and ineffective means to measure impact that lack a deep understanding of the returns. Amount of funds raised, growth in the number of participants, number of supported projects, and so on, are often used as indicators of success, but such statistics are problematic.
For instance, participants of a platform are often “dormant”, meaning they register but do not use the platform later on. It is necessary to change the way “social impact” is understood by policy makers and investors to distinguish what needs to be measured and what not, and if measurement is a must the focus should be on tangible changes that the platform brings. For example, which regulations have changed as a result of platform activities? Which medical research results are obtained by patient-doctor platforms? Which civic projects are realised, and what are potential benefits? Social indicators should focus on a deeper understanding of how the actual social practices that give rise to social problems are tackled, and what the role of platforms are in this process.
8. Innovation (un)readiness of population
While most of the policy focus is on supporting the generation of innovations, the innovation readiness of the user population is not given enough attention. Investments in developing Internet skills are of crucial importance, which include operational, formal and strategic skills. The research of Alexander Van Deursen and Jan Van Dijk provides insight on this question.
In addition, potential users can be unaware, uninterested, or unconnected even if they have a benefit to gain. Paradoxically, those who are most likely to benefit from DSI are more likely to be unaware, uninterested, or unconnected. Instead of being confined to the online sphere, social entrepreneurs should work actively with target populations in the field, in developing solutions and encouraging participation. As Tom Saunders of NESTA states, it is important to “remember that there’s a world beyond the Internet”. For example, the city of Amsterdam is remarkable in efforts to integrate the people in the collaborative economy.
9. Duplication, duplication, duplication
Most digital platforms operate according to the logic of network externalities, also called as multi-sided platforms. This means that the existence of one group of users in a platform makes it more attractive for other groups to join. In this way, certain digital platforms build up their user base rapidly and become dominant players. While this can be problematic in terms of building up of monopolistic power, too many start-ups in the same field is also problematic, which is the case today in some areas of DSI.
For example, there are more than 20 civic-tech platforms with similar functions in France. The potential gains and losses in terms of social welfare and efficiency should be understood and evaluated better in the case of DSI. Many of these platforms struggle to grow, their user base is divided, and finally they close down within a few years of launching. One solution can be to allow for sharing reputation, or other information about users between platforms, which helps in sustaining diversity, while avoiding centralisation of power.
10. Lack of cross-fertilisation
The importance of the above problems also depends on the field of activity and type of DSI considered, as there are many different types of DSIs. Aggregating all DSIs in a single group may be misleading. At the same time it is precisely this diversity that gives this emerging ecosystem its dynamism and resilience. Unfortunately, this diversity is not made use of in an effective way. Instead, field-specific bubbles have formed with weak interactions between them. Cross-fertilisation and synergies between these are potentially important to increase resilience, but networks rest weak. A recent initiative in France is Plateformes en Communs, which aims to form a common platform of cooperatives and associations in diverse domains of activity, so as to leverage synergies between them.
Given the high level of penetration of digital technologies in our lives, digital social innovations promise to address big challenges, yet for there to be better outcomes, more needs to be done. Participation to civic life – online or offline – is always valuable in an increasingly problematic world. Digital platforms make this participation much easier. As the saying goes, little drops of water make a mighty ocean.
[divider style=”dotted” top=”20″ bottom=”20″]
DSI4EU, Muge Ozman and Cedric Gossart are organising a special stream on digital social innovations in the 10th International Social Innovation Conference, which will take place in Heidelberg, in September 2018.
https://imtech-test.imt.fr/wp-content/uploads/2018/03/Visuel_Une_980x1344.jpg9791344I'MTechI'MTech2018-03-23 13:55:252018-05-11 11:40:03Can digital social innovations tackle big challenges?
The second edition of the IMT–Académie des Sciences Awards is now open for applications. These awards strive to reward exceptional scientific contributions at the European level in three areas: science and technology relating to digital transformation in industry, science and technology relating to energy transition, and environmental engineering. The deadline for applications is April 24, 2018.
Two scientists honored
2017 Awards Ceremony at the Académie des Sciences
The IMT–Académie des Sciences Awards comprise two awards:
– a Grand Prix awarded to a scientist who has made an exceptional contribution to the fields mentioned above through an outstanding body of work;
– a Young Scientist Prize awarded to a scientist who is under 40 years old on January 1 of the year the prize is awarded, and who has contributed to these same fields with a major innovation.
These prizes will be awarded jointly by IMT, with support from Fondation Mines-Télécom, and the Académie des Sciences. They will include the following prize amounts:
– Grand Prix: €30,000
– Young Scientist prize: €15,000
Each prize will be awarded to a scientist of any nationality working in France, or in Europe in close collaboration with French team.
Applications must include:
1) the form provided by the Académie des Sciences
2) a letter of support providing a personal opinion of the candidate
3) a short curriculum vitae
4) the candidate’s main scientific results
5) a list of the main publications
Official awards ceremony
The formal awards ceremony will be held in the dome of the Academy in Autumn 2018. It will be accompanied by a ceremony for the winners to present their work to the Academy.
https://imtech-test.imt.fr/wp-content/uploads/2018/03/Institut_France_2.jpg8901309I'MTechI'MTech2018-03-22 14:34:582019-01-09 10:18:572018 Edition of the IMT–Académie des Sciences Awards
The startup Vizity, incubating at ParisTech Entrepreneurs, seeks to reinvent how content is shared online. It has developed a mapping solution that combines online resources related to a place, making them more easily accessible for users.
Timothée Lairet, Co-founder of Vizity
“To talk about places, nothing beats a map,” Timothée Lairet assures us. By reminding us of this often-forgotten truth, the young entrepreneur sums up the purpose behind Vizity, the startup he cofounded. Because what better way to combine resources about cities from blogs, online travel guides, city hall and tourist offices than with a map? This is exactly what the startup proposes to do, “We gather these different types of content and combine them on a map to make them more accessible,” Timothée Lairet explains.
For now, the startup incubated at ParisTech Entrepreneurs works with each stakeholder independently. When working with the tourist office for a city or region, for example, it first creates a map of the geographical area, which will be added to the organization’s website. Then, Vizity can dynamically link an article from the tourist office’s website to an area on the map. A blogpost about an exhibition at a museum, or the history of a castle will be displayed when a site visitor explores these areas on the map.
The solution addresses a problem faced by tourists who do are not familiar with an area. “They know the information is out there, but don’t know how to look for it,” the co-founder explains. It is hard to find information about a village market, for example, if you don’t even know the market exists. But with the map, the site user sees the event on the map as a point of interest. By clicking on the linked content, the tourist can find opening hours for the market and what vendors will be there, and then decide whether or not to go.
Besides tourists, residents of big cities can also benefit from this solution. In a city inhabited by hundreds of thousands or even millions of people, it’s easy to miss out on an event we’re interested in that took place just a few minutes from home. Blogs that offer ideas for outings would benefit from having their latest updates included on a map that would be open to users. This would ensure we never miss the information added to our neighborhood map.
Thanks to Vizity maps, the different producers of content about a place, whether it be bloggers, companies or administration services, can offer their own view of the city’s places of interest and share it with others. By combining informational content for each site, they offer unique content curation and can recommend original tour ideas to their customers and users.
Towards a new form of map media?
The startup’s long-term goal is to offer a comprehensive map that would be open to users and bring together different types of content for the same location. Information on an exhibition at a prestigious museum, historical information about the museum’s building, and a review for the associated gourmet restaurant would all be available on the same map, even though the content would be from different websites.
Tanguy Abel, Co-founder of Vizity
And while the idea of combining a map and reviews could make you think of Google Maps, the comparison stops there. For Timothée Lairet, the goal is not to produce another review aggregator, but to focus on content with a high added value for users, written by professionals or a circle of close friends. The map must allow users to access a wealth of valid and valuable information.
By pairing this solution with the geolocation of users, Vizity also hopes to offer new services. For a start, tourist offices could better understand their visitors’ behaviors and better meet their needs, making their stay more enjoyable. More importantly, based on past visits, users could determine options for their next visit, and save their preferences. When travelling abroad, we would just need to tell the Vizity app what we’re looking for, and it would propose a visit that matches the experience we want. In short, a new way to explore locations off the beaten track.
https://imtech-test.imt.fr/wp-content/uploads/2018/03/Visuel_Une_English.jpg9521400I'MTechI'MTech2018-03-21 11:02:302018-03-21 11:02:30Vizity: explore the city with digital maps
Starting with Algorithm and ending with Virus, this list features terms like Phishing and Firewall… As the symposium entitled “Are we entering a new era of cybersecurity?” is getting underway at IMT, here are 24 words to help you understand the concepts, technologies and systems used to protect people, materials and organizations from cyberattacks. This glossary was compiled with the help of Hervé Debar, a researcher at Télécom SudParis, an expert in cybersecurity and co-organizer of the symposium.
Algorithm – A sequence of instructions intended to produce a result (output data) by means of a calculation applied to input data.
Critical infrastructures– Infrastructures for which a cyberattack could have very serious consequences for the services provided, even to the point of putting lives at risk.
Cryptography– The science of secrets. Cryptography proposes algorithms that can make data unreadable for those who do not have the secret. It also makes it possible to sign digital documents.
Cyberattack– A sequence of actions that lead to the violation of the security policy. This violation often takes the form of a computer system or network malfunction (inability to connect, a service that is no longer available, or data being encrypted using ransomware). A cyberattack can also be invisible, but lead to serious consequences, such as the theft of confidential information.
Cyber defense – A country’s means of attacking and defending its computer systems and networks.
Cyber range– A training platform for cyberattacks and defense.
Denial of Service Attack (see Distributed Denial of Service Attack)
Distributed Denial of Service Attack (DDoS Attack) – An attack aimed at overloading a service provider’s resources (often related to the network), making it inaccessible.
Electromagnetic injection – An electromagnetic signal sent to disrupt the operation of an electronic component (processor, memory, chip card…).
Firewall – A network component that filters incoming and outgoing traffic on a website.
Flaw– A (software) flaw is a programming error made by the programmer that allows a hacker to run a program for a different use than what was intended. The most prevalent example is SQL injection, in which hackers use a web site’s interface to control databases they could not normally access.
Krack(Key Reinstallation Attacks) – Attacks against the WPA2 protocol that allow an attacker to force the reuse of an encryption key. This allows the attacker to collect a large number of packets, and therefore decrypt the network traffic more easily, without knowing the key.
Malicious software(seeMalware)
Malware– A program used for a purpose that is inconsistent with the user’s expectations and violates the security policy. Malware often uses vulnerabilities to enter a system.
Phishing – A social engineering technique, in which an attacker convinces a victim to act without understanding the consequences. The technique often relies on emails with fraudulent content (e.g. CEO fraud scams).
Ransomware– Malicious software (malware) aimed at extorting money from a victim, often by encrypting the data on their computer’s hard disk and demanding payment in exchange for the decryption key (often these keys are useless, and purchasing them is therefore useless).
Resilience (by design) or cyber-resilience– A system’s ability to function in the event of an attack, that is, provide a service to its users in any condition, albeit at a reduced level.
Security Information and Event Management – A platform for uploading and processing alerts that allows operators to monitor their systems’ security and react in the event of an attack.
Trojan Horse– A backdoor installed on a system without the users’ and administrators’ knowledge, which allows a hacker to regularly and easily connect to the system without being seen.
Virus – Malicious software capable of entering a system and spreading to infect other systems.
https://imtech-test.imt.fr/wp-content/uploads/2017/11/Visuel_Une_900x1350.jpg8971352I'MTechI'MTech2018-03-20 11:31:382018-03-21 11:54:1924 words for understanding cybersecurity
Several terabytes: this is the phenomenal amount of data produced by the Sentinel satellites each day! How can these data flows be used to develop concrete applications to be used by those who manage territories? This is what spatial application experts focused on at the AppSpace Forum, an event organized by the CNES, GIS BreTel, Booster Morespace and Institut InSpace from October 17 to 19, 2017.
Copernicus, a program run by the European Space Agency and the European Union, has launched the Sentinel satellites – 1A, 1B, 2A, 2B and 3A, each equipped with different sensors for taking a variety of measurements. The goal is to provide European users, and more specifically researchers, with comprehensive, free observational data of the entire Earth: oceans, land, vegetation, coastal areas, radiometry, temperature, altimetry, etc. But can spatial data be used to develop concrete applications?
The question of an application for data from the Copernicus program was at the forefront of the AppSpace event, co-organized by GIS Bretel. For the first time, this initiative brought together all actors in the region of Brittany, but also spatial application professionals from throughout France and Europe, to participate in round table discussions, themed workshops, and an exhibition space for companies and laboratories. The organizers of Appspace intend it to become a reference event taken up by other regions of France and Europe. The goal is to obtain a clear, overall vision of the regional, national, and even European ecosystem of spatial apps.
Encourage end users to take possession of data
“The takeaway from this event is that, generally speaking, the world of research is quite good at taking possession of spatial data”, explains Nicolas Bellec, operational director of GIS BreTel. “However, some field specialists, such as biologists or ecologists, sometimes have difficulty using these data in their research, and call on other labs specialized in the field of space.”
Beyond the world of research, data from the Copernicus program were especially designed to help territorial authorities and regional and State services, to meet their own needs. But these actors, considered as the end users, do not use these data. “But the resolutions of the new satellite sensors are increasingly well-adapted to their needs!” states Nicolas Bellec. “The scope of applications is also very broad: maritime safety and security, land use and regional planning, monitoring of vegetation and biodiversity, adaptation to climate change, etc. At the Appspace forum, we tried to understand why.”
By bringing together the worlds of research, companies and end users, the Appspace event highlighted the barriers to using spatial data and finding appropriate solutions. “What we found was that territorial managers lack training and information on these subjects. Researchers, companies and users want to create applications together, to better meet the needs of territorial managers” Nicolas Bellec explains.
The other difficulty is that spatial data can rarely be the sole solution to a concrete problem. They often must be used with other data, and in particular, field data, to find their place in applications. There are several ongoing projects which manage to incorporate spatial data into existing processes of information acquisition in the territories.
The Sésame project, created by Lab-STICC*, the teams Obelix and Myriads from IRISA and funded by the DGA and the ANR, crosses spatial data with AIS data from ships to develop applications for monitoring and surveillance of maritime traffic.
Develop technologies capable of handling data flows
The goal of the Sésame project is to develop technologies capable of detecting and giving real time documentation of unusual behavior of ships: illegal entries into defined areas, suspicious deviations from trajectories, illegal fishing, etc. To achieve such a result, high resolution photographs of the water’s surface produced by Sentinel satellites need to be crossed with AIS (Automatic Identification System) data emitted by ships. Each ship emits an AIS signal, which includes information on the ship itself, its route and position, at resolutions per minute. The challenge of the project is to process these extremely high flows of data. On top of the terabytes of data from the Sentinel satellite, tens of millions of AIS messages are produced each day.
“CLS, our industrial partner, is a solutions operator for monitoring maritime traffic using satellite data. The current data processing chains will need to be reviewed in order to cope with the scale of the flows which are currently being produced” explains Ronan Fablet, professor and researcher in the Lab-STICC laboratory at IMT Atlantique and coordinator of the Sesame project. “The company is embarking on research and development processes to use Big Data and Machine Learning technology in monitoring maritime activity. The Sésame project is an integral part of this process.” With a consortium of teams specialized in Big Data, Machine Learning and remote detection, the goal of Sésame is to manage data flows with the development of suitable material and software infrastructure, and to develop machine learning techniques for detecting ships and unusual behavior in satellite images.
These technological developments are intended to be used first of all by CLS, then made available to operators such as the ESMA, the institution in charge of surveillance of European maritime areas. “Overall, the end users targeted by the project are the institutions in charge of maritime surveillance for regions, states, or groups of states” specifies Ronan Fablet.
Finally, as well as offering solutions to concrete issues of maritime traffic surveillance, the technologies developed by the Sésame project will pave the way for the use of already existing databases, by associating them with other types of satellite imagery. With the development of adapted infrastructure and Big Data technology, the gigantic data flows produced by Sentinel satellites will also be channeled, processed and interpreted, to serve the development of many other applications designed for end users.
* Members of Lab-STICC: IMT Atlantique, UBO, UBS, CNRS, ENIB and ENSTA Bretagne
https://imtech-test.imt.fr/wp-content/uploads/2017/11/DKt9qJHXkAAIkQH.jpg-large.jpg17952048I'MTechI'MTech2018-03-16 16:48:532018-03-16 16:48:53What are the applications for spatial data?
What is energy? What is power? Roland Lehoucq, an astrophysicist at CEA and professor at École Polytechnique and Sciences Po, uses Science Fiction to help explain scientific principles to the general public. Star Wars, Interstellar, The Martian… These well-known, popular movies can become springboards for making science accessible to everyone. During his conference on “Energy, Science and Fiction” on December 7th at IMT Mines Albi, Roland Lehoucq explained his approach to popularizing science.
What approach are you taking for this conference on “Energy, Science and Fiction”? How do you use science fiction to study science?
The goal of this conference is to use science fiction as a springboard for talking to the general public about science. I chose the cross-cutting theme of energy and used several science fiction books and movies to talk about the topic. This drives us to ask questions about the world we live in: what prevents us from doing the things we see in science fiction? This question serves as a starting point for looking at scientific facts: explaining what energy and power are, providing some of the properties and orders of magnitude, etc. In general, the fictional situations involve levels of energy and power that are so significant that, for now, they are beyond our reach. Humanity does have a great deal of energy within its grasp, which is why it has been able to radically transform planet Earth. But will this abundance of energy last? Will we someday reach the levels we see in science fiction? I’m not so sure!
My approach is actually the same as that of science fiction. It dramatizes scientific and technical progress and is designed to make us think about the consequences of these developments. This can apply to energy, genetics, artificial intelligence, robots, etc. It involves questioning reality, but it has no qualms about distorting the facts make a more appealing story. Works of fiction pay no attention to significant scientific facts, choosing to happily ignore certain physical laws, yet this is not truly a problem. It does not affect the works’ narrative quality, nor does it change the questions they raise!
Does this type of approach allow you to reach a wider audience? Do you see this at your speaking events?
I don’t know if I am reaching a wider audience, but I do see that those in the audience, both young and old, are delighted to talk about these subjects. I use some of the best-known films, although they are not necessarily the most interesting ones from a scientific point of view. While Star Wars does not feature a lot of high-level thinking, it is nevertheless full of content, including energy, which can be analyzed scientifically. For example, we can estimate the Jedis’ power in terms of watts and rank them. My approach is then to say: let’s imagine this really exists, let’s look at the information we can draw from the film and, in return, what we can learn about our world. Young people respond positively since I use things that are part of their culture. But it works well with other generations too!
What led you to share scientific culture using science fiction as the starting point?
I have loved science since I was 6 years old. I started reading science fiction when I was 13. Then I taught about science as a group leader at astronomy camps from the age of 17 to 23. I have always enjoyed learning things and then talking about the aspects I find to be the most interesting, amazing and wonderful! It comes naturally to me!
Then, in the early 2000s, I decided I wanted to share my knowledge on a larger scale, through books and articles. I quickly got idea of using fictional literature, comic strips and the cinema as a way of sharing knowledge. Especially since no one was doing it then! If you want to talk about astrophysics, for example, you have people like Hubert Reeves, Michel Cassé, Marc Lachièze-Rey and Jean-Pierre Luminet who are making this knowledge accessible. I did not want to repeat what they were already doing so well. I wanted to break away and do something different adapted to my tastes!
What advice would you give to researchers on improving how they share scientific culture?
Sharing scientific knowledge with others is not intuitive for researchers because it essentially involves making difficult choices of only saying what is most useful for the general public in a limited amount of time. Often researchers focus their life work, intelligence and efforts on a very limited topic. Of course, researchers will want to talk about this area of expertise. But to understand the reasons that led the researcher to work in this area, the audience first needs certain prerequisites. And if these prerequisites are not provided, or are incomplete, the audience cannot understand the interest of the subject and the issues being discussed. It is therefore necessary to take the time to explain what researchers see as general information. Therefore, for one hour of a conference, forty-five minutes must be spent presenting the prerequisites and fifteen minutes spent explaining the field of research. This requires making a choice to serve the field, to take a back seat and avoid the “specialist syndrome”, which involves talking only about what the specialist sees as important, their 10 or 15 years of research. This is a legitimate approach, but by doing this researchers risk losing their audience!
They must also try to make science “friendly”. Science is often seen as something complicated, which requires great effort to be understood. As is often the case, a lot of work is needed to understand the subtleties of these subjects. Our job therefore consists of facilitating access to these areas, and the methods chosen will depend on each individual’s interests. Finally, we must show the general public that science is not an accumulation of knowledge, but an intellectual process, a methodology. We can therefore study science as an educational exercise, using things that are not purely scientific, such as science fiction!
[box type=”shadow” align=”” class=”” width=””]
Roland Lehoucq
Associate Professor of Physics and former student of ENS, Roland Lehoucq is an astrophysicist at the CEA center at Paris-Saclay, and teaches at the Ecole Polytechnique and at the Institut d’Etudes Politiques de Paris. He has written numerous books for the popularization of science using the science fiction as the starting point, such as La SF sous les feux de la science and Faire de la Science avec Star Wars. He recently wrote a book on the dark ideas of physics, Les idées noires de la physique, published by Les belles lettres, in collaboration with Vincent Bontems, a philosopher of science, and illustrated by Scott Pennors. Black holes, dark matter, dark energy… This book looks at all these subjects through the eyes of an astrophysicist and a philosopher.
[/box]
https://imtech-test.imt.fr/wp-content/uploads/2018/03/Visuel_Une_carré.jpg9441132I'MTechI'MTech2018-03-15 11:27:062019-08-09 12:33:00When Science Fiction Helps Popularize Science – An Interview with Roland Lehoucq
The driving force behind the modernization of the electrical network, the smart grid is full of promise. It will mean savings for consumers and energy companies alike. It terms of the environment, it provides a solution for developing renewable energies. Hossam Afifi, a researcher in networks at Télécom SudParis gives us a behind-the-scenes look at the smart grid.
What is the purpose of a smart grid?
Hossam Afifi: The idea behind a smart grid is to create savings by using a more intelligent electric network. The final objective is to avoid wasting energy, ensuring that each watt produced is used. We must first understand that today, the network is often run by electro-mechanical equipment that dates back to the 1960s. For the sake of simplicity, we will say it is controlled by engineers who use switches to remotely turn on or off the means of production and supply neighborhoods with energy. With the smart grid, all these tasks will be computerized. This is done in two steps. First, by introducing a measuring capacity using connected sensors and the Internet of Things. The control aspect is then added through machine learning to intelligently run the networks based on the data obtained via sensors, without any human intervention.
Can you give us some concrete examples of what the smart grid can do?
HA: One concrete example is the reduction of energy bills for cities, municipal authorities and hence local taxes, and major infrastructures. A more noticeable example is the architectural projects for buildings that feature both offices and housing, which are aimed at evening out the amount of power consumed over the course of the day and limiting excessive peaks in energy consumption during high-demand hours. The smart grid rethinks the way cities are designed. For example, business areas are not at all advantageous for energy suppliers. They require a lot of energy over short periods of time, especially between 5pm and 7pm. This requires generators to be used to ensure the same quality of service during these peak hours. And having to turn them on and off represents costs. The ideal solution would be to even out the use of energy, making it easier to optimize the service provided. This is how the smart grid dovetails with smart city issues.
The smart grid is also presented as a solution to environmental problems. How are the two related?
HA: There is something very important we must understand: energy is difficult to store. This is one of the limits we face in the deployment of renewable energies, since solar, wind and marine energy sometimes produce electricity at times when we don’t need it. However, a network that can intelligently manage the energy production and distribution is beneficial for renewable energies. For example, electric car batteries can be used to store the energy produced by renewable sources. During peaks in consumption, users can choose to disconnect from the conventional network and use the energy stored by their car in the garage and receive financial compensation from their supplier. This is only possible with an intelligent network that can adapt the offer in real time based on large amounts of data on production and consumption.
How important is data in the deployment of smart grids?
HA: It is one of the most important aspects, of course. All of the network’s intelligence relies on data; it is what feeds the machine learning algorithms. This aspect alone requires support provided by research projects. We have submitted one proposal to the Saclay federation of municipalities, for example. We propose to establish data banks to collect data on production and consumption in that area. Open data is an important aspect of smart grid development.
What are the barriers to smart grid deployment?
HA: One of the biggest barriers is that of standardization. The smart grid concept came from the United States, where the objective is entirely different. The main concern there is to interconnect state networks, which up until now were independent, in order to prevent black-outs. In Europe, we drew on this concept to complement the deployment of renewable energies and energy savings. However, we also need to interconnect with other European states. And unlike the United States, we do not have the same network standards as our German and Italian neighbors. This means we have a lot of work to do at a European level to define common data formats and protocols. We are contributing to this work through our SEAS project led by EDF.
https://imtech-test.imt.fr/wp-content/uploads/2018/03/ile-de-nantes.jpg12801920I'MTechI'MTech2018-03-14 12:35:382022-03-18 17:11:48What is a smart grid?
DSI4EU, Muge Ozman and Cedric Gossart are organising a special stream on digital social innovations in the 
Roland Lehoucq
