The researchers present a novel approach that enables photo-realistic re-animation of portrait videos using only an input video.
While existing methods use only manipulations of facial expressions, they are the first to transfer the full 3D head position, head rotation, face expression, eye gaze, and eye blinking from a source actor to a portrait video of a target actor.
AutoAugment could be useful to extend the image data set used for training. Instead of manually rotation and flipping images, it should be possible to be possible to figure out the way how to increase both the amount and diversity of data in an existing training dataset with AutoAugment.
AutoAugment is an automatic way to design custom data augmentation policies for computer vision datasets, e.g., guiding the selection of basic image transformation operations, such as flipping an image horizontally/vertically, rotating an image, changing the colour of an image, etc.
Recently Google has been releasing many machine learning examples of machine and Seedbank. Seedbank is a registry and search engine for Colab notebooks for and around machine learning for rapid exploration and learning.