Face de-identification has become increasingly important as the image sources are explosively growing and easily accessible. The advance of new face recognition techniques also arises peoples concern regarding the privacy leakage. The mainstream pipelines of face de-identification are mostly based on the k-same framework, which bears critiques of low effectiveness and poor visual quality. In this paper, we propose a new framework called Privacy-Protective-GAN (PP-GAN) that adapts GAN with novel verificator and regulator modules specially designed for the face de-identification problem to ensure generating deidentified output with retained structure similarity according to a single input. We evaluate the proposed approach in terms of privacy protection, utility preservation, and structure similarity. Our approach not only outperforms existing face de-identification techniques but also provides a practical framework of adapting GAN with priors of domain knowledge.
Flexible photonics is rapidly emerging as a promising platform for artificial smart skins to imitate or extend the capabilities of human skins. Organic material systems provide a promising avenue to directly fabricate large-scale flexible device units; however, the versatile fabrication of all-organic integrated devices with desired photonic functionalities remains a great challenge. Here, we develop an effective technique for the mass processing of organic microlaser arrays, which act as sensing units, on the chip of photonic skins. With a bilayer electron-beam direct writing method, we fabricated flexible mechanical sensor networks composed of coupled-cavity single-mode laser sources on pliable polymer substrates. These microlaser-based mechanical sensor chips were subsequently used to recognize hand gestures, showing great potential for artificial skin applications. This work represents a substantial advance toward scalable construction of high-performance and low-cost flexible photonic chips, thus paving the way for the implementation of smart photonic skins into practical applications.
Circularly polarized light (CPL) detection has emerged as a key technology for various optoelectronics. Chiral hybrid perovskites (CHPs) that combine CPL-sensitive absorption induced by chiral organic ligands and superior photoelectric properties of perovskites are promising candidates for direct CPL detection. To date, most of the CHP detectors are made up of polycrystalline thin-film, which results in a rather limited discrimination of CPL due to the existence of redundant impurities and intrinsic defect states originating from rapid crystallization process. Here, it is developed a direct CPL detector with high photocurrent and polarization selectivity based on low-defect CHP single-crystal nanowire arrays. Large-scale CHP nanowires are obtained through a micropillar template-assisted capillary-bridge rise approach. Thanks to the high crystallinity and ordered crystallographic alignment of these arrays, a CPL photodetector with high light on/off ratio of 1.8 × 10 4 , excellent responsivity of 1.4 A W −1 , and an outstanding anisotropy factor of 0.24 for photocurrent has been achieved. These results would provide useful enlightenment for direct CPL detection in high-performance chiral optoelectronics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.