Our products are used to:
_reduce the uncertainty of Machine Learning models,
_prioritize enterprise data efforts,
_support experts in the ML loop,
_improve the quality of ML models, especially in multi-class settings with complex ontologies,
_reduce data footprint and compactify ML models so as to be used by the Internet of Things applications,
_improve gaming experience via more challenging and realistic AI in games,
_create intelligent advisory systems from pre-compiled building blocks.
BlindBox is a technology that allows for irreversible anonymisation of visual data (images and video streams), while maintaining the ability to analyse it in order to extract information relevant to the user, but without being able to identify the persons whose images are contained in the data. BlindBox enables enterprises to obtain information supplementing their knowledge of their clients/publicity and translate it into specific actions, e.g., in the area of marketing or advertising campaign profiling.
- A method for anonymizing (distorting) the original data and its hardware implementation of the Field-Programmable Gate Array (FPGA) in a way that makes it irreversibly impossible to interpret for people and standard computerized image analysis tools.
- Algorithms/analytic models dedicated to such "distorted" data, that will be able to extract information valuable from the client's point of view (only to the extent permitted by legal regulations), including historical analyses;
- Implementation in real-life conditions, e.g. reducing the volume of processed data and limiting the required computing power.