Despite the great progress made by deep CNNs in image semantic segmentation, they typically require a large number of densely-annotated images for training and are difficult to generalize to unseen object categories. Few-shot segmentation has thus been developed to learn to perform segmentation from only a few annotated examples. In this paper, we tackle the challenging few-shot segmentation problem from a metric learning perspective and present PANet, a novel prototype alignment network to better utilize the information of the support set. Our PANet learns classspecific prototype representations from a few support images within an embedding space and then performs segmentation over the query images through matching each pixel to the learned prototypes. With non-parametric metric learning, PANet offers high-quality prototypes that are representative for each semantic class and meanwhile discriminative for different classes. Moreover, PANet introduces a prototype alignment regularization between support and query. With this, PANet fully exploits knowledge from the support and provides better generalization on few-shot segmentation. Significantly, our model achieves the mIoU score of 48.1% and 55.7% on PASCAL-5 i for 1-shot and 5-shot settings respectively, surpassing the state-of-the-art method by 1.8% and 8.6%.
Dielectric capacitors have become a key enabling technology for electronics and electrical systems. Although great strides have been made in the development of ferroelectric ceramic and thin films for capacitors, much less attention has been given to preventing polarization fatigue, while improving the energy density, of ferroelectrics. Here superior capacitive properties and outstanding stability are reported over 107 charge/discharge cycles and a wide temperature range of −60 to 200 °C of ferroelectric Aurivillius phase Bi3.25La0.75Ti3O12‐BiFeO3 (BLT‐BFO), which represents one of the best capacitive performances recorded for the ferroelectric materials. The modification of BLT thin films with BFO overcomes the constraints of ferroelectric Aurivillius compounds and presents an unprecedented combination of the ideal features including improved polarization, reduced ferroelectric hysteresis, and lowered leakage current for high‐energy‐density capacitors. Given the lead‐free and fatigue‐free nature of this Aurivillius phase ferroelectric, this work unveils a new approach towards high‐performance eco‐friendly ferroelectric materials for electrical energy storage applications.
Texture features indexed by mean and GLN based on the arterial phase images reflect biologic aggressiveness, and may have potential applications in predicting the histological grading of HCC preoperatively. Evidence level: 4 J. MAGN. RESON. IMAGING 2017;45:1476-1484.
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.