Collaborative Tool

One of the project goals is to develop a low barrier interactive tool to allow interview subjects to fine tune their own level of abstraction. It should be noted that our multistage AI art process is not an image processing process — it does not change the pixels of the frames of the source regular or 360 video (as you might do in an Adobe product). Instead, it specially paints a painting style result of every frame of the video. It is an open and dynamic process, allowing levels of control throughout. It is our eventual goal that either the interviewee or documentary producer can use options in the system to customize the final result based on their wants and needs. The following is a more detailed description of the process.

The first stage is a two-step process that uses AI to note facial features of the subject, and intelligently change the features before the painting process. This is the first level of anonymization. Our system can identify the regions of the face (as they are moving dynamically in a regular or 360 video) and can automatically change many features including the width of the eyes, the separation between the eyes, the height and width of the nose, the size of the forehead, the size of the cheeks and chin, and the overall shape of the head.

 

Different level of abstraction of the Picasso-like cubist representation of the subject:


 

 


 

Line edges and two forms of AI painter processing are added:


 

This multi-step process insures that it can not be easily reversed (since it does not use just one AI technique and it has random steps) and more so emulates the fine art painterly process to best represent the subject — stylistically in a way humans appreciate (and will hopefully engage with longer and more intently) and in a way the can best convey the subject’s thoughts and feelings even through some level of abstraction/distortion. Because it is dynamic and customizable, the subject or the producer can have control over the level of anonymization (depending on their needs) as well as make adjustments to adjust the emotion or tone conveyed.