Convolutional neural networks (CNNs) have been widely used in computer vision community, and significantly improving the state-of-the-art. How to train an intra-class variant and inter-class discriminative feature is a central topic in face recognition. This paper proposes to learn an effective feature from face images by a joint loss function which combines the hard sample triplet (HST) and the absolute constraint triplet (ACT) loss, under the criteria that a maximum intra-class distance should be smaller than any inter-class distance. With the joint supervision of HST and ACT loss, CNNs is enable to learn discriminative features to improve face recognition performance. Experiments on labeled faces in the wild, IARPA Janus Benchmark (IJB-A) and YouTube Faces datasets achieve a comparable or superior performance to the state-of-the-arts.
Obtaining a delamination‐free wrinkled functional graphene surface in layered systems is an interesting challenge because the interface is usually too weak to withstand interfacial stress mismatch, which can trigger mechanical instability. In this paper, a general strategy is proposed toward addressing the delamination limitation imposed by fabricating conformal graphene wrinkles with bilayer systems of poly(methyl methacrylate) (PMMA) and polydimethylsiloxane (PDMS). To improve the interfacial strength, a postcuring transfer process is introduced to form a gradient interface layer without interfacial liquid between the PMMA and PDMS by entanglement of polymer chains during high‐temperature curing. Compared to the conventional wet transfer of graphene,the transfer method can greatly enhance the interfacial strength. The chemical and mechanical mechanisms underlying the enhancement are revealed both experimentally and theoretically in terms of the transition from the buckled‐induced delamination state to the delamination‐free wrinkled state. Moreover, the light diffraction behaviors of multiscale graphene wrinkles are initially demonstrated to be an interesting continuous pattern induced by overlapping. The delamination‐free conformal wrinkled functional graphene surface can provide valuable insight and design guidelines for the fundamental problems of deformed graphene and its applications in flexible functional devices. 相似文献
