How does the blur feature operate? How does the software identify individuals' faces and where this data is stored?

For the facial blur, the software records videos in 2D and then transforms the object into a 3D model of a skeleton. Then the software identifies the face, blurs it out, and saves this version of the video to the cloud. The new version of the video with the blur gets saved in the cloud, whereas the original version of the video is deleted from the device simultaneously and can’t be restored.

For the background blur, TuMeke creates the 3D model of a skeleton and tracks the model throughout the video, not the person. The new version of the video with the blur gets saved in the cloud, whereas the original version of the video is deleted from the device simultaneously and can’t be restored.