Photo-realistic 360 Head Avatars in the Wild
This addresses the challenge of making immersive 3D human communication accessible using only commodity hardware, though it is incremental in improving pose estimation for avatar creation.
The paper tackles the problem of creating 360-degree photo-realistic head avatars from mobile phone videos by proposing a novel landmark detector trained on synthetic data to estimate camera poses, enabling a multi-stage optimization process that produces realistic avatars from any viewpoint.
Delivering immersive, 3D experiences for human communication requires a method to obtain 360 degree photo-realistic avatars of humans. To make these experiences accessible to all, only commodity hardware, like mobile phone cameras, should be necessary to capture the data needed for avatar creation. For avatars to be rendered realistically from any viewpoint, we require training images and camera poses from all angles. However, we cannot rely on there being trackable features in the foreground or background of all images for use in estimating poses, especially from the side or back of the head. To overcome this, we propose a novel landmark detector trained on synthetic data to estimate camera poses from 360 degree mobile phone videos of a human head for use in a multi-stage optimization process which creates a photo-realistic avatar. We perform validation experiments with synthetic data and showcase our method on 360 degree avatars trained from mobile phone videos.