Video processing for Equestrian competition videos
Horse-racing video processing application
The application is created for equestrian organizers. It's a Windows desktop application with complex machine learning algorithms on the backend. Mostly, OpenCV library is used. For better results we improved the algorithms for our particular purpose. It's only required to have videos from different cameras and a list of racers.
The interface of the application allows setting folders where the videos are located on the computer, upload a .xls file with a list of racers (or a Google Sheets link). Nothing else is required. All the magic happens behind users view: AI models are trained, videos are processed. It is possible to turn off certain riders to reduce processing time.
Riders and horses detection.
As a result, the application returns Rider's name, Horse colour, Rider's number, and the most important - Time interval for this particular Rider. All this data can be recorded to Google Sheets document or can be downloaded as a .xls file.
Also, it's possible to generate a video with custom music on it containing all the videos from different cameras for this particular Rider detected by the algorithm. We've reached about 90% of accuracy by training Models on different video arrays
Complex solution for oil consumers and suppliers
The platform for consumers and suppliers of oil and petrochemicals to set a feasible connection between them.
Virtual try-on tool for makeup products
The system consists of a face detection and segmentation model and an algorithm that allows recoloring objects without losing their original texture.
Online sign language interpreter
AI algorithm that converts video of a person using sign language into a text transcript