GPU-Based Embedded Intelligence Architectures and Applications

This paper present contributions to the state-of-the art for graphics processing unit (GPU-based) embedded intelligence (EI) research for architectures and applications. This paper gives a comprehensive review and representative studies of the emerging and current paradigms for GPU-based EI with the focus on the architecture, technologies and applications: (1) First, the overview and classifications of GPU-based EI research are presented to give the full spectrum in this area that also serves as a concise summary of the scope of the paper; (2) Second, various architecture technologies for GPU-based deep learning techniques and applications are discussed in detail; and (3) Third, various architecture technologies for machine learning techniques and applications are discussed. This paper aims to give useful insights for the research area and motivate researchers towards the development of GPU-based EI for practical deployment and applications.

Embedded Intelligence on FPGA: Survey, Applications and Challenges

Embedded intelligence (EI) is an emerging research field and has the objective to incorporate machine learning algorithms and intelligent decision-making capabilities into mobile and embedded devices or systems. There are several challenges to be addressed to realize efficient EI implementations in hardware such as the need for: (1) high computational processing; (2) low power consumption (or high energy efficiency); and (3) scalability to accommodate different network sizes and topologies. In recent years, an emerging hardware technology which has demonstrated strong potential and capabilities for EI implementations is the FPGA (field programmable gate array) technology. This paper presents an overview and review of embedded intelligence on FPGA with a focus on applications, platforms and challenges. There are four main classification and thematic descriptors which are reviewed and discussed in this paper for EI: (1) EI techniques including machine learning and neural networks, deep learning, expert systems, fuzzy intelligence, swarm intelligence, self-organizing map (SOM) and extreme learning; (2) applications for EI including object detection and recognition, indoor localization and surveillance monitoring, and other EI applications; (3) hardware and platforms for EI; and (4) challenges for EI. The paper aims to introduce interested researchers to this area and motivate the development of practical FPGA solutions for EI deployment.

Novel artificial intelligence-driven software significantly shortens the time required for annotation in computer vision projects

Background and study aims The contribution of artificial intelligence (AI) to endoscopy is rapidly expanding. Accurate labelling of source data (video frames) remains the rate-limiting step for such projects and is a painstaking, cost-inefficient, time-consuming process. A novel software platform, Cord Vision (CdV) allows automated annotation based on “embedded intelligence.” The user manually labels a representative proportion of frames in a section of video (typically 5 %), to create ‘micro-modelsʼ which allow accurate propagation of the label throughout the remaining video frames. This could drastically reduce the time required for annotation. Methods We conducted a comparative study with an open-source labelling platform (CVAT) to determine speed and accuracy of labelling. Results Across 5 users, CdV resulted in a significant increase in labelling performance (P < 0.001) compared to CVAT for bounding box placement. Conclusions This advance represents a valuable first step in AI-image analysis projects.

Motion-core assistive tools using pervasive embedded intelligence

Real-time human movement monitoring anywhere at any time is time critical depending on core human motion activities, in particular nation’s valuable asserts; athletes and soldiers considered as reference standard of any society. Light weight wearable technologies are the key measurements and instruments system integrated to develop human motion-core assistive tools (MAT) using pervasive embedded intelligence. Unlike many existing motion analysis models, motion-core models are based on domain specific data service architectures beyond cloud technologies using inner data structures and data models created. Four layered micro system architecture that consists of sensing, networking, service and Motion-core IoT (MIoT) is proposed. Knowledge base was designed as a distributed and networked data center based on transient and resident data addressing modes in order to guarantee the secure data accessing, propagating, visualizing and control between these two modes of operations. While transient data change and avail in relevant clouds storages, corresponding resident data and processed data retain inside local servers or/and private clouds. Data mapping and translation techniques are applied for the formation of complete motion-core data packet related to the test subject under consideration. Thus, hybrid MIoT system is developed using 3D decision fusion models which are the internationally quantifiable standards for assessing human motion set by trainers, coachers, physiotherapists and orthopedics. MIoT built as motion-core assistive tools have been tested for rehabilitation monitoring, injury prevention and performance optimization of athletes, soldiers, and general public. The hybrid system introduced in this work is novel and proves lower down the latency and connectivity independence by allowing human movement analysis during daily active lifestyle.

Simulator with Embedded Intelligence Focused on the Design Process

Mechatronic design practice was conceived as various successive steps involving expertise. However, employers expect recently graduated engineers to start working with the shortest training period. This paper reports a research that developed a simulation tool that introduces modifications and additions to the regular methods of dynamic simulation, integrating in it several of the steps of the systematic mechanic design. The design tool encompasses for each element or object of the simulator seven new features that in an intelligent way gives the student a little design practical expertise. The connection between elements follows the method by wires and the window of assembly includes a workspace where the 3D depiction of all elements is seen and animated according to the values of the variables. The concept was prototyped and now all technological components are available to start the development of a product. This concept, due to the volume of information that it uses, instead of being attractive to cover all fields of knowledge is valuable to adapt to specific fields as academic courses. Potential users evaluated the attractiveness of the concept through a work section, giving good indicators.