Generally, facial expressions could be classified into two categories: static facial expressions and micro-expressions. There are many promising applications of facial expression recognition, such as pain detection, lie detection, and babysitting. Traditional convolutional neural network (CNN)-based methods suffer from two critical problems when they are adopted to recognize micro-expressions. First, they are usually dependent on very deep architectures that overfit on small datasets. However, reliable expressions are relatively difficult to collect and relevant datasets are usually relatively small. Second, for micro-expressions, these methods usually neglect the temporal redundancy of micro-expressions which could be utilized to reduce the temporal complexity. In this paper, we propose a shallow CNN (SHCNN) architecture with only three layers to classify static expressions and micro-expressions simultaneously without big training datasets. To better explain the functionality of our SHCNN architecture, we improve the saliency maps by introducing a shrinkage factor after studying the vanishing gradient problem of existing saliency maps. Experiments are conducted on five open datasets: FER2013, FERPlus, CASME, CASME II, and SAMM. To the best of our knowledge, by comparing with other methods offering source code (or pseudo code), we believe that our method would be the best on FERPlus, CASME, and CASME II and competitive on FER2013 and SAMM.
Accelerated growth of urban population in the world put incremental stresses on metropolitan cities. Smart city centric strategies are expected to comprise solutions to sustainable environment and urban life. Acting as an indispensable role in smart city, IoT (Internet of Things) connects the executive ability of the physical world and the intelligence of the computational world, aiming to enlarge the capabilities of things in real city and strengthen the practicality of functions in cyber world. One of the important application areas of IoT in cities is food industry. Municipality governors are withstanding all kinds of food safety issues and enduring the hardest time ever due to the lack of sufficient guidance and supervision. IoT systems help to monitor, analyze, and manage the real food industry in cities. In this paper, a smart sensor data collection strategy for IoT is proposed, which would improve the efficiency and accuracy of provenance with the minimized size of data set at the same time. We then present algorithms of tracing contamination source and back tracking potential infected food in the markets. Our strategy and algorithms are evaluated with a comprehensive evaluation case of this IoT system, which shows that this system performs well even with big data as well.
Chip multiprocessor (CMP) techniques have been implemented in embedded systems due to tremendous computation requirements. Three-dimension (3D) CMP architecture has been studied recently for integrating more functionalities and providing higher performance. The high temperature on chip is a critical issue for the 3D architecture. In this article, we propose an online thermal prediction model for 3D chips. Using this model, we propose novel task scheduling algorithms based on rotation scheduling to reduce the peak temperature on chip. We consider data dependencies, especially inter-iteration dependencies that are not well considered in most of the current thermal-aware task scheduling algorithms. Our simulation results show that our algorithms can efficiently reduce the peak temperature up to 8.1 • C.
We present a analytical framework to identify the tradeoffs and performance impacts associated with different SoC platform configurations in the specific context of implementing multimedia applications. "Configurations" in this case might include sizes of different on-chip buffers and scheduling mechanisms (or associated parameters) implemented on the different processing elements of the platform. Identifying such tradeoffs is difficult because of the bursty nature of on-chip traffic arising out of multimedia processing and the high variability in their execution requirements, which result in a highly irregular design space. We show that this irregularity in the design space can be precisely captured using an abstraction called variability characterization curves.
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.