Occluded Face Recognition Based on Deep Image Prior and Robust Markov Random Field
The occlusion-caused difference between test and training images is one of the most challenging issues for real-world face recognition system.Most of the existing occluded face recognition methods based on deep neural networks(DNNs)need to use large-scale occluded face images to train network models.However,any external object in the real world might become occlu-sions,and limited training data cannot exhaust all possible objects.Also,using large-scale occluded face images to train networks violates the human visual mechanism,the human eyes detect occlusions by only using small-scale unoccluded face images without seeing any occlusions.In order to simulate the occlusion detection mechanism of human vision,we combine the deep image prior with the robust Markov random field model to construct a novel occlusion detection model,namely DIP-rMRF,based on small-scale data,and propose a uniform zero filling method to effectively utilize the occlusion detection results of DIP-rMRF.Experi-mental results of six advanced DNN-based face recognitions methods,including VGGFace,LCNN,PCANet,SphereFace,Inter-pretFR and FROM,on three face datasets,including Extended Yale B,AR and LFW,show that DIP-rMRF can effectively prepro-cess face images with occlusions and quasi-occlusions caused by extreme illuminations,and greatly improve the performance of the existing DNN models for face recognition with occlusion.
Face recognition with occlusionDeep image priorRobust Markov random fieldUniform zero-fillingStructural error metric
NETL
NSTL
万方数据
