Exposit

OMN is a powerful online tool for solving various tasks in the following areas: Product information management, Project management, Media asset management, Brand management, Workflow management and Media production management. The main modules with powerful functionality are combined and configured in accordance with the processes of a specific customer in a single and complete system of media production and large companies multi-channel marketing management.

Our task was to finalize the product, customize it for specific customers as well as to develop specific plug-ins.

While working on this software product, Exposit gained unique experience in the geographically distributed development of a large software product (about 300 person-years, more than 70 of which are Exposit), release and branch management, analysis, automation and implementation of complex production cycles and processes for a lot of the end customers, such as McDonalds, IKEA, Meyle & Muller.

Functions of the OMN application:

• Automation of media content production (for example, pdf-catalogs of high resolution);
• Media logistics;
• Product information management.

Technologies: J2ЕЕ, Maven, Spring, Hibernate, Tomcat, XML, Webservices, Red5 Server, Active MQ, Adobe Flex, JS/CSS, NFS Helios, Oracle, Quark Xpress Server, Adobe InDesign Server, ImageMagick, Ghostscript, FFmpeg, ImgSeek, Adobe XMP SDK, HTML2PDF

VR visualization of the Great Synagogue of Slonim for the Belarus Shtetl project aimed at the actualization of Jewish heritage. 
 

Task: The Second World War, the Holocaust, mass emigration, and Soviet ideology had a devastating effect on the culture of Belarusian Jews: Jewish heritage and its role in the history of Belarus and Europe still remain unknown to modern residents of the region. That's why the Belarusian-Jewish Cultural Heritage Center with the support of the Goethe-Institut Minsk decided to revive the Jewish heritage using VR technologies.
 

Solution: Exposit implemented VR visualization of the Great Synagogue of Slonim for an interactive exhibition.
 

Our team began the development by visiting the historical building in Slonim to better understand the architecture and create a 3D model of the synagogue as realistic as possible. 
 

We explored the real object, saw elevation plans and historical materials, took photographs of interior and exterior details as well as made photographic materials for creating a 360-view inside and outside the synagogue.
 

The preparation of high-quality digital materials for detailing and the use of Unity 3D allowed us to create a realistic 3D model as the basis for a virtual tour of the Great Synagogue of Slonim. Users can plunge into the architectural context wearing VR glasses and freely move around the building. 

Functionality:

• Creation of a 3D model of the synagogue in its original form and 360-view of the current state of the building;
• Support for a VR headset without controllers including the ability to interact with elements by holding the viewer gaze;
• Support for a VR headset with controllers (only motion control by pressing buttons) without interaction with the elements by hands;
• Implementation of a synagogue tour in a virtual environment based on a custom script (movement through points of interest, musical accompaniment).

Technologies: Unity3D, SteamVR, Blender3D

A multi genre Unity game with a beagle as a main character.

Wow Dog is a dog owner simulator that allows you to treat, feed, wash the dog and play with it, as well as to take part in online exhibitions. You can participate in Wow Dog Show to show your clever, educated and beloved dog around the world. The main goal of the game is to raise the best dog in the world and win every competition.

Task: Unity game development from scratch, manual testing. 

Solution: Our team has developed a virtual Unity-based pet game from scratch. It is a pet owner simulator that allows you to care of a beagle. All the implemented functionality has been covered by manual tests including functional (black box testing), localization and usability testing.

Implemented functionality:

• Ability to care for a dog (you can wash, feed, put to sleep and play);
• Exhibition mini game - ability to take part in competitions and earn coins;
• Different levels of difficulty;
• Ability to spend earned coins on the new clothes.

Planned functionality:

• Wow Dog Selfie that allows its user to photograph the best moments of the game and share them with friends in social networks;
• Wow Dog Chat that allows user to communicate with the dog on any topics. Users can write different questions and receive funny answers sharing stickers, stories and compliments. Answers can be shared with friends.

Technologies: Unity3D

arterio is a custom mobile solution for heart patients. This application allows users to monitor blood pressure and view key measurement statistics in one place. It helps patients observe health dynamics and share the latest measurement data with their doctors. This way patients can timely seek treatment at medical centers and get the most appropriate medication.

Task: More than half of all strokes and heart attacks are caused by high blood pressure. It is also a risk factor for heart diseases, kidney diseases, and vascular dementia. Only about 24% of U.S. adults have their hypertension under control while measuring blood pressure on a regular basis is an essential step in preventing high blood pressure. Hypertension treatment and control rates in Europe are even lower.

Access to regular blood pressure readings can help heart patients avoid risks by making more informed decisions on medicine prescriptions required for hypertension management. However, around 75% of clinics use manual paper logs for their patients that contain reporting bias affecting up to 50% of all BP self-measurements. Exposit carefully studied the problem of blood pressure monitoring and decided to develop a mobile software solution that would provide heart patients with a convenient way of making timely log measurements and reduce the risks of heart attacks.

Solution: Our Team developed a Healthcare solution designed to meet the specific needs of heart patients and make their lives easier and safer. Our app allows recording, monitoring, managing, and sharing blood pressure measurements data using easy navigation.

Technologies: Bluetooth, Kotlin, MVVM, Material 2.0, DataBinding, Room, LiveData

MEDsafe is a custom web-based system designed for doctors and their patients. The system provides a convenient way of keeping full health-related information in one place and retrieves patients’ data from mobile and clinical devices, wearables, and connected applications.

Task: Development of a solution that will meet the needs of healthcare providers and their patients to improve the way clinics store and exchange medical data.

Solution: Our healthcare specialists developed a health-record web app from scratch that helps to store, manage and share medical measurements within a single platform. This platform is easy to use for both doctors and their patients providing communication on treatments and care plans between two parties.

This system provides authorization via Google and Dutch DIGID for easy and secure access to sensitive data. Doctors and their patients can create, store and manage medical measurements, care plans, and other health-related data but there is also the admin role for system management. If users want to discuss care plans together with their families and doctors, they can invite them to their network. Thanks to compliance with HL7 standard users can easily share and retrieve medical data from other healthcare organizations and services like Google Fit or Apple healthkit. It helps to improve the interoperability between different care providers and healthcare information systems. 

MEDsafe is a safe and secure EHR and EMR application thanks to FHIR based storage and limited access to users’ data. The project is Information security management certified (ISO 27001), Dutch information security management standard certified (NEN 7510), MedMij certified, and also certified for Medical Devices (ISO 13485).

Technologies: Angular 7, TypeScript, Redux, Sass, Java11, SpringBoot 2.1, Spring Framework, Docker, PostgreSQL, HAPI HL7 FHIR, Kubernetes, Zuul

Football Analytics is a computer vision system designed to improve the quality of training analytics and in-club competitions. 

Task:

1. Analyze the overall feasibility of a CV system to improve the training analysis.
2. Development of a software solution solving specific tasks of offline football schools.

Solution: Our team worked on a soft-and-hardware solution that allows users to track the movement of football players and balls on the field and use the collected data to create individual training programs. During the development, we took into account cost optimization in terms of equipment, processes automation, and a large amount of video data.

We reviewed the existing options for sports games analytics to create a solution that meets all the requirements: wearable tracking systems and optical tracking systems.

During the research, we found that these tracking systems are often used in conjunction for greater accuracy of football analytics. Thus, we created a solution that combines the capabilities of wearable and optical systems to collect and process various types of data without using expensive equipment. This solution turned out to be a computer vision system. 

To test the solution, our team developed a prototype of football analytics based on CV technologies, as well as an operating algorithm: from calibrating cameras and ground marking to receiving a statistical report. The prototype can detect the player’s position on the football field during a match or training session and identify a specific player using the “reference” histograms. Histograms graphically illustrate the number of pixels at each color intensity level creating a unique graph for each player. The result of the system is a file containing data on the coordinates of the players.

Technologies: OpenCV, Tensorflow, Keras, Mask R-CNN, Deep SORT, Open Pose

Wizart is a neural network-based computer vision and augmented reality solution for visualizing finishing materials and creating interior designs. The product provides customers with a unique opportunity of trying out wallpapers, painting, and flooring in your own home before buying. You can instantly see the new materials in your own interior with the help of a simple photo: you just turn on a phone camera, select a suitable type and material and get a photorealistic result preserving the perspectives, scale, and shadows. 

Task: Development of CV Interior Assistant solving the problem of the imagination gap.

Solution: Exposit took part in the development of the own product Wizart including an AI-driven iOS application and Web application that can be easily integrated with an E-Commerce website. Together we created a solution that uses a neural network to show customers new interior design ideas in just a few clicks. After receiving a photo, the application recognizes the ceiling, floor, furnishings, and decorations. Then, users have a chance to apply their favorite finishing materials options on the photo and the existing materials are replaced by a new choice. Users can try different types and colours until they satisfied with the results and finished the room repair. We also added an Augmented Reality mode designed for a more accurate overlay.

At the end of the project, we developed an iOS mobile application with a working algorithm and a web application embedded in E-Commerce websites. After the project completion, Wizart became a separated company bringing competitive advantages to retailers and unleashing customers' creativity. 

Integration options:

• A plug-in for E-Commerce websites that allows working with app as a with regular online store;
• A workspace or self-service terminal for the offline store.

Technologies: Swift, Python, Keras, Tensorflow, CoreML, Numpy, cv2, OpenCV, ARkit, Docker

Learn more: https://www.exposit.com/portfolio/wizart/