Facelet-Bank for Fast Portrait Manipulation

Researchers from The Chinese University of Hong Kong, Tencent, and Johns Hopkins University have developed a tool that automatically adds, removes, or modifies facial features of a person in an image. It is based on an end-to-end convolutional neural network that supports fast inference, edit-effect control, and quick partial-model update. In addition, this model learns from unpaired image sets with different attributes. As an example, it can add face hair to your picture, or make you look younger or older.

Courtesy of the researchers

The work has been accepted by the Conference on Computer Vision and Pattern Recognition (CVPR) 2018, which will be held in Salt Lake City, Utah from June 18-22.

The full paper is available at https://arxiv.org/pdf/1803.05576.pdf.