Although the facial makeup transfer network has achieved high-quality performance in generating perceptually pleasing makeup images, its capability is still restricted by the massive computation and storage of the network architecture. We address this issue by compressing facial makeup transfer networks with collaborative distillation and kernel decomposition. The main idea of collaborative distillation is underpinned by a finding that the encoder-decoder pairs construct an exclusive collaborative relationship, which is regarded as a new kind of knowledge for low-level vision tasks. For kernel decomposition, we apply the depth-wise separation of convolutional kernels to build a light-weighted Convolutional Neural Network (CNN) from the original network. Extensive experiments show the effectiveness of the compression method when applied to the state-of-the-art facial makeup transfer network -- BeautyGAN.

You might be familiar with the word Artificial intelligence. You might have heard on the radio or might have seen some news stories or a Hollywood movie depicting the same. And why all of the sudden there is such a trend about it in recent times. For starters, the thing is so popular that almost all the top companies from around the world, who are in the FORTUNE 500 are partially or directly involved in evolving or developing or using some form of ARTIFICIAL intelligence. Be it, Tesla, Facebook, Google, Microsoft, OpenAI, and the list goes on and on. Artificial intelligence (AI), generally called machine intelligence, is intelligence determined by machines.