Security&Safety Things AppChallenge
The year 2020: a year of change, but also of new beginnings and innovations. Many things have changed already, and many more will likely change in the future. For us at TNG, the time spent in seclusion has been a valuable opportunity to reflect, but also to create something new!
The Security and Safety Things App Challenge 2020 provided the perfect setting to demonstrate our computer vision skills under time pressure. In just three months we managed to create two innovative apps – all the way from idea to development to availability in the marketplace. Not only that but both apps were actually awarded first prize in their categories: Visual Feedback won in the Retail and Commerce category and Dynamic Privacy Mask in the Smart Cities category!
This year we not only wave goodbye to handshakes, but also to unsanitary feedback buttons! As retail store managers told us, customer feedback given via physical, in-store feedback buttons has indeed decreased dramatically. We remedied this with our app Visual Feedback! Visual Feedback enables customers to give feedback with a simple hand gesture, e.g. a thumbs up or thumbs down (demo video). To make the process even more playful and engaging, emojis that correspond to the given feedback are then displayed in place of the customer's head. Giving feedback with Visual Feedback is not only contactless, but also fun!
Our second app is a bit more serious. Many of us feel uncomfortable when high-resolution security cameras record personal data. Of course this cannot be completely avoided (after all, cameras are there to record people!) but that does not mean that credit cards or PIN pads have to be filmed at the grocery store or that the screen of your laptop needs to be captured in the waiting area at the airport. In fact, GDPR even demands that the collection of personal data be limited to what is necessary ("data minimization"). With Dynamic Privacy Mask, the recording of personal data can now be minimized even further. Using a Neutral Network, our app masks object classes such as people, laptops, and keyboards (demo video) – and soon also PIN pads, credit cards, and mobile phones. Through our app we hope to raise the standard of data protection and privacy in the long term. After all, GDPR also states that operators of security cameras must adhere to the technological state of the art when it comes to protecting personal data (and, in particular, everyone's right to informational self-determination).
Want to learn more about our apps? Drop us a line at iot-apps(at)tngtech.com for more information and let us know what you think!
Hey oven, open up!
Talking to your household appliances sounds like Science Fiction, right? Well not anymore! Just as Ali Baba once did to open the door to his enchanted treasure trove, the owners of a new oven can now call out "Open Sesame" to open it. A simple "Hey Alexa, open the oven“ would also suffice.
This fairytale come true won our customer 1st Place for "Excellence in Business to Consumer“ at this year's German Innovation Awards. The voice support was developed jointly by TNG and our customer. Now when you have your hands full, using a voice-assistant like Amazon Alexa, you can simply ask the oven to open the door for you.
We are delighted to be a part of the innovations of the future.
TNG demonstration App featured in S&ST Product Demo
Our demonstration app Person Anonymizer featured in a live online product demo for the S&ST platform. Cameras running the OS developed by Security and Safety Things can execute machine learning algorithms on-device rather than in the cloud. Such cameras can perform a wide range of complex detection tasks without having to send any video data over the wire.
The App showcases how people caught on CCTV cameras can be detected and masked in real time, protecting their privacy whenever surveillance of individuals is not the camera's intended use case. The video is available on YouTube.
Page Gardener App helps keeping Confluence content up to date
Often on company wikis, information is written once and then quickly forgotten about. This information is then neither updated nor archived, even though it may be out of date. An app designed by TNG, our Page Gardener Companion for Confluence, helps address this issue.
Users and administrators can define responsible persons who will be reminded at regular intervals to maintain and update the contents of particular sites and spaces within Atlassian Confluence. This automation helps keep your pages up to date, avoids misunderstandings that cost time and money, and allows you to focus on the content, not the process.
Have we piqued your interest? Then check out the Atlassian Marketplace and try the app for 30 days completely free.
TNG goes virtual
Sometimes, we have to make a virtue of necessity. In view of the spreading SARS-CoV-2 virus TNG started an experiment last Friday, a “Virtual Techday.” This consisted of virtual workshops, presentations and even a virtual “Weekly half-hour,” an informative in-house event hosted by the TNG partners. The Weekly Half-hour had previously been streamed, in addition to the actual presentation. However this was the first time as a purely virtual video conference. Up to 230 colleagues attended simultaneously, applauded via emoticons, and posted comments in the chat.
The feedback on the virtual Techday experiment was extremely positive. However, we are all looking forward to meeting again in person.
Remedy for Performance Issues
The colleagues will talk about this topic at the Virtual Big Techday on May 8th, 2020.
Black Friday: every year this top-selling day poses a unique challenge to our rapidly growing customer in the online retail industry. The webshop must operate reliably under an unprecedented load. Despite a hardware upgrade, in 2019 there were already short service disruptions observed at peak times prior to Black Friday. Trouble was clearly imminent.
We identified and resolved multiple performance bottlenecks. The configuration of the application was then optimised in an iterative fashion. This was done using a JMeter test plan with request distribution modelled on customer behaviour. Executing it from the cloud directly against the production servers enabled us to test different load scenarios. At the same time, technical and organisational measures were taken to ensure that customer shopping remained undisturbed. Elastic Stack, Prometheus & Grafana served as the main tools to identify and analyse backend metrics.
Ultimately the improvements yielded the desired result. The throughput doubled, but the system withstood this crucial phase without any issues. The new monitoring processes allows future performance problems to be detected, understood, and rapidly solved.
Deep Fakes in real time
Deep Fakes are imitations of pictures and video, created using artificial intelligence. Common scenarios include exchanging faces in order to create the illusion to see another person in a video. Within the scope of a research project, the TNG Innovation Hacking team investigated Deep Fakes to better understand what is technically feasible. Furthermore, we wanted to explore the limitations of the technology - especially if and with what image quality it is possible to create Deep Fakes in real time.
Our research shows that it is possibled to exchange the face of a person filmed via a real-time video streamwith the face of another person, including applying facial expressions and movements of the person filmed.
By applying various techniques from the area of computer vision and neural networks, faces in the video feed are recognized, transformed and embedded in the video output. The project uses autoencoder networks trained in Keras. They were trained using so called GANs (Generative Adversarial Networks). Aditionally, the developers used different neuronal networks for face recognition and segmentation.