Cross domain meta-network for sketch face recognition

Because of the large modal difference between sketch image and optical image, problem that traditional deep learning methods are easy to overfit in case a small amount training data, Cross Domain Meta-Network for face recognition method is proposed. This first designs meta-learning strategy solve sample problem, then proposes entropy average loss cross domain adaptive reduce domain. The experimental results on UoM-SGFS PRIP-VSGC data sets show this other methods.