top of page

Face Anonymization Tool

Location: Germany

Industry: Well-being

Face anonymization tool designed to empowering individuals to comfortably share their traumas with psychologists, ensuring their privacy and peace of mind by safeguarding their identity from recognition.

 

The tool should offer the capability to create a new facial appearance by choosing various features from library such as the nose, cheeks, beards, and eyebrows, allowing users to seamlessly replace their original face with this generated version.

Знімок екрана 2023-09-14 о 13.45.jpg

Challenges:

The primary objective in tool development was to enable real-time face swapping on the user's device, ensuring that the raw video stream remains on the device without external transmission.

 

Another significant challenge involved creating a feature-rich library for generating realistic new faces. This functionality had to be seamlessly integrated into the tool, affording users complete control over combining various features without any limitations.

Job Done:

We designed a Face generator that can work with unlimited quantity of faces features that can be downloaded from the library and create any face from it.

 

Also, we designed face swap tool that used for work more than 400 landmark points on user face by neural network, for providing the most accurate face merging and fully transfer user emotion. This tool can work on user device in real time and do not require using a powerful graphical card for real time video processing.

Implemented Features

deep learning counter drone system

The collision of birds and drones with airplanes is a huge issue for ensuring flight safety. Radars and other warning systems cannot always definitively identify a target and classify it. For the final classification of the target, it is worth using optical systems, and artificial intelligence allows you to relieve the operator and automatically classify the object.

The collision of birds and drones with airplanes is a huge issue for ensuring flight safety. Radars and other warning systems cannot always definitively identify a target and classify it. For the final classification of the target, it is worth using optical systems, and artificial intelligence allows you to relieve the operator and automatically classify the object.

deep learning counter drone system

Outcomes:

Runway inspection is a mandatory procedure for the airport. Personnel must check the runway for defects and objects that could damage the aircraft. Even the presence of small garbage on the runway can lead to catastrophic consequences. Visual inspection by airport personnel always carries the risk of missing important details due to the human factor, at the same time, computer vision systems allow not only improve the quality of the inspection but also speed it up.

Client Feedback

Runway inspection is a mandatory procedure for the airport. Personnel must check the runway for defects and objects that could damage the aircraft. Even the presence of small garbage on the runway can lead to catastrophic consequences. Visual inspection by airport personnel always carries the risk of missing important details due to the human factor, at the same time, computer vision systems allow not only improve the quality of the inspection but also speed it up.

bottom of page