We are experts in high-performance computing system design and customization. Whether cloud, edge, embedded or mix-of-all, we develop custom full-stack solutions that meet your application’s performance and power requirements. With over 15 cumulative years of experience in commercial and defense related projects, our experts are qualified and committed to finding your unique solution.
We are hardware and software engineers with intimate knowledge of high-performance system design
Is your application missing performance targets? Does your application exceed the energy budget of your embedded or cloud platform? In the era of Artificial Intelligence the “magic” behind the most compelling applications are complex and computationally intense algorithms. Often the compute requirements of these algorithms far exceed the capabilities of general purpose computing. For these increasingly common scenarios, SiliconScapes develops and implements hardware accelerated solutions customized for those critical algorithms. The result? Maximum raw performance and/or performance per watt for your application.
Whether a proof-of-concept embedded SoC prototype or ready to deploy cloud appliance, we’ve got you covered. Developing FPGA-based prototypes and demonstrations was how we got our start and it remains a core expertise.
System Specification Authoring
Need help going from requirements to specification? Let us help you define the ideal platform and components to realize your application.
Ready to revamp your engineering team? New hire, small team, large team, onsite, offsite, or remote… Our training services will get your team to proficiency.
What we are good at
High Performance Hardware Design
Low Power Design
Custom Data Processing Pipelines
Algorithm Analysis & Optimization
Numerical Requirements Analysis
System Bus Interfacing
A key task in image analytics is identifying regions-of-interest (ROI). ROI are areas in an image that are likely to contain objects of importance to the application. Once identified, ROI can be subsequently analyzed to detect the presence of these objects of importance. By focusing object recognition only on ROI, high system throughput can be maintained without loss of overall scene understanding functionality. This accelerator solution provides real-time region-of-interest detection capabilities for machine vision and video analytics applications. This accelerator would typically by the first stage of a visual object detection and classification system. This solution was demonstrated at OpenPOWER 2016 and SuperComputing 2016
FPGA-based Hardware Acceleration of SURF Algorithm
Many machine vision algorithms use visual features for object detection. At a high-level, an object’s visual features represent a unique fingerprint that can be used to distinguish and identify the object in cluttered scenes. When searching an image for a particular object, the unique fingerprint is matched against all fingerprints in the image. SURF is a classical visual feature that has many applications in object detection, image registration, augmented reality and object tracking. Our FPGA-based accelerator computes SURF features for HD imagery while maintaining real-time frame rate of greater than 30 Frames Per Second.
Real-time Vision with FPGA + IBM POWER8/9
Complex vision algorithms often require computing power beyond the capabilities of general purpose servers. This limits the degree of natural human interaction that can be supported by the application. To overcome this, hardware accelerators are designed specifically for each key algorithm to achieve interactive performance. These accelerators are tightly integrated into the IBM POWER8/9 with CAPI for high performance and scaleability.
Our clients & affiliates
2171 Sandy Drive, Suite 205, State College, Pa, 16803