Dev kit for creating low-power FPGA smart embedded vision solutions

21-05-2020 | Microchip Technology | Development Boards

With the rise of AI, ML and the IoT, applications are shifting to the network edge where data is collected, demanding power-efficient solutions to provide more computational performance in even smaller, thermally constrained form factors. Through its Smart Embedded Vision initiative, Microchip Technology is answering the increasing demand for power-efficient inferencing in edge applications by making it simpler for software developers to perform their algorithms in PolarFire FPGAs. As a notable addition to the solutions portfolio in this segment, the company's VectorBlox Accelerator SDK assists developers in taking advantage of its FPGAs for producing low-power, flexible overlay-based neural network applications without learning an FPGA tool flow.

The highly flexible tool kit can produce models in TensorFlow and the ONNX format, which provides the widest framework interoperability. ONNX supports many frameworks, including Caffe2, MXNet, PyTorch, and MATLAB. Unlike alternative FPGA solutions, the company's VectorBlox Accelerator SDK is supported on Linux and Windows operating systems, and it also incorporates a bit accurate simulator which gives the user the chance to validate the accuracy of the hardware while in the software environment. The neural network IP included in the kit also supports the capability to load various network models at run time.

“In order for software developers to benefit from the power efficiencies of FPGAs, we need to remove the impediment of them having to learn new FPGA architectures and proprietary tool flows, while giving them the flexibility to port multi-framework and multi-network solutions,” said Bruce Weyer, vice president of the Field Programmable Gate Array business unit at Microchip.

“Microchip’s VectorBlox Accelerator SDK and neural network IP core will give both software and hardware developers a way to implement an extremely flexible overlay convolutional neural network architecture on PolarFire FPGAs, from which they can then more easily construct and implement their AI-enabled edge systems that have best-in-class form factors, thermals and power characteristics.”

Related product news