Under the hood copernic360 is based on artificial intelligence (AI) techniques to work with 360° photos and videos. 360° photos and videos are simply examples of data defined on the sphere.

Spherical data-sets are in fact very common and are found not only in VR but in a wide variety of fields, including geophysics, astrophysics, medical imaging, communication systems, acoustics, and computer vision and graphics, to mention just a few.

360° VR. (Source)

Earth. (Source)

Cosmic microwave background relic light from the Big Bang. (Source)

White matter fibres in the brain imaged by diffusion MRI. (Source)

Computer graphics mesh. (Source)

All of the examples shown above often involve the analysis of spherical data.

We’ve been hard at work generalising the spherical AI techniques behind copernic360.

Today we’re annoucing our fourpiAI platform for AI for spherical data.

fourpiAI dramatically outperforms standard approaches to apply AI to spherical data. On a common benchmark classification problem involving spherical data, standard methods achieve an accuracy of only 50%, whereas fourpiAI achieves an accuracy of 90%.

fourpiAI transfers the success of deep learning to spherical settings, unlocking the remarkable potential of deep learning to solve a wide range of new problems in VR and beyond.

To learn more about fourpiAI check out our related product page or get in touch!