Become an anime character, a cartoon character, or even a classical painting with this new tool.
Guoxian Song, Linjie Luo, Jing Liu, Wan-Chun Ma, Chunpong Lai, Chuanxia Zheng, and Tat-Jen Cham from Nanyang Technological University and ByteDance have revealed a great new tool that can create stylized portraits by inversion-consistent transfer learning – AgileGAN. This tool allows you to upload your photo, preferably with a neutral expression, and get a stylized portrait of yourself like in the examples below.
"We introduce a novel hierarchical variational autoencoder to ensure the inverse mapped distribution conforms to the original latent Gaussian distribution while augmenting the original space to a multi-resolution latent space so as to better encode different levels of detail," states the team. "To better capture attribute-dependent stylization of facial features, we also present an attribute-aware generator and adopt an early stopping strategy to avoid overfitting small training datasets."