Discover our products
Software products we have developed. We work hard and we are dedicated to producing the best possible outcome, as you can see from our case studies.
Sales and upsell prediction modeling
Our model analyzes customer databases, purchase history, renewal history, and CLSV to identify customers most likely to purchase or upgrade. This reduces churn, shortens the sales funnel, and accelerates the upgrade process. Additionally, we examined the factors that influenced past purchases to replicate the same success with new customers and encourage repeat purchases from existing ones. By analyzing available data, we built predictive models that help shorten conversion times and increase conversion rates.
Generating 3D Animation for Games, AR & VR Solutions
Solution for automatic generation of 3D objects animation for various purposes including AR & VR applications and 3D games. It allows to reproduce a wide variety of natural-looking virtual objects like cars, characters, etc, including interactions between a characters and an objects. The main benefit of this solution is that through automation we drastically simplify and lower cost of 3D objects and character animation while supporting wide range of object modification.
Product recognition for vending machines/smart fridges
Computer vision based app for real-time detection and recognition of products from video. The solution can be used in stores to transform small self-serving shops into fully automated Amazon Go-like cashierless stores. It can help food delivery services track dispatched orders. For brick and mortar stores it can help streamline the purchasing process, reduces shoplifting, and cut costs on RFID tags.
Ecommerce recommendation engine
The system delivers accurate product recommendations, enhancing upsell and cross-sell opportunities without relying on generic external systems. It is powered by an ensemble of trained machine learning models, including collaborative filtering, natural language processing with deep recurrent neural networks, and image processing using deep convolutional neural networks.
_gif.gif)
Retail Traffic Analysis
Visitor analytics solutions that enable physical stores to collect and analyze in-store footfall data. We've utilized Computer Vision and AI to develop models that help brick-and-mortar retailers optimize store space and leverage customer insights through camera-based people counting. Key features include people counting, visitor path mapping, heatmap generation (crowd density, dwell time, overall traffic), indoor wayfinding analysis, queue monitoring and management, as well as MAG (mood/age/gender) visitor profiling.
_gif.gif)
AI-powered attention heatmap
Perceptbox is a visual content analytics platform that effectively predicts human attention and attraction. It simulates human vision during the first few seconds of exposure to visuals and generates an attention heatmap using AI algorithms that predict what a real person would most likely focus on. Perceptbox identifies which elements of visual content capture attention and which are overlooked. The application is written in C++, utilizing OpenCV and artificial neural networks.
_gif.gif)
Social distance measurement
Our model can measure the distance between pedestrians in public spaces. Using a traffic light indicator system, the algorithm anonymously identifies and labels individuals who maintain safe distances, while flagging instances in red where social distancing measures are violated. This information helps to identify bottlenecks where social distancing cannot be maintained and provides insights into how citizens adapt as restrictions are imposed or lifted.
_gif.gif)
Gender, age, emotion detection
We've used TensorFlow and OpenCV, trained on the IMDB-2013 and UTKFace datasets to create a custom model. This model can also be applied to eye color detection projects, as it accurately extracts eyes from any given image. By leveraging this technology, businesses can enhance facial recognition systems, improve user authentication processes and offer personalized user experiences.
_gif.gif)
Hair segmentation
We developed this solution using a PyTorch model based on U-Net with a MobileNet v2 backend for precise hair segmentation, trained on the Figaro1k dataset. The model delivers highly accurate results, making it ideal for applications such as virtual try-ons, image editing, and augmented reality. Its lightweight design ensures fast processing, making it perfect for both mobile and web platforms, enhancing user experience and enabling seamless integration into beauty and fashion tech solutions.
Data extraction from invoices, forms & other unstructured documents
We've developed a data extraction tool that retrieves information in key-value format, transforming documents into business-ready data optimized for processing, analysis, and storage. This tool streamlines data management by automating manual extraction processes, reducing errors, and saving time. It enables businesses to quickly access structured data for more efficient decision-making, improving workflow efficiency, and enhancing the accuracy of downstream analytics and reporting.
Let's start our cooperation!
