Ximilar offers consultancy, customization, development, and software deployment in similarity management of contemporary digital data collections. It finds its applications in content-based management of complex objects such as biometric characteristics, events in computer or social networks, image, video, and music retrieval, time series of various nature, and many others.

The offered technology is based on principles of:

  • extensibility: applicable on a wide variety of data types and forms of similarity (e.g. digital images, videos, biometrics, music, various scientific data, etc.) - see application case studies
  • scalability: efficient even for very large data collections (billions of objects)
  • HW independence: using infrastructure as a service, various centralized or distributed hardware infrastructures can be applied thus performance can be tuned to meet requirements of specific applications

On conceptual level, the technology offered by Ximilar adopts the metric space paradigm explained in a fundamental book and experimentally studied in the MUFIN project. On the software level, the technology capitalise on the open source code.

Feel free to contact us with any questions.