German researchers at the Max Plank Institute have developed a high-level software tool that allows users to exchange or animate faces in images in an almost completely automated way.
The new software tool first reconstructs a 3D face model from the original image, applies changes to it in 3D, and draws the result into a target image.
The new pose and illumination for the target image is computed automatically.
The only manual interaction required by the user is to click on about seven feature points, such as the corners of the eyes and mouth, in each image.
The original image of the woman’s face (left) was automatically inserted into the two consecutive pictures using the new software.
The researchers envisage their new software being used in a variety of photo-editing applications – it could, for example, be used in consumer applications where a user might wish to see how he or she looked with a different hairstyle.
And, they add, extending the technique to video frames is straight-forward.