FPGA-based Neuromorphic Accelerator board recognizes objects 7x more efficiently than GPUs on GoogleNet, AlexNet

[ ]

BrainChip Holdings has just announced the BrainChip Accelerator, a PCIe server-accelerator card that simultaneously processes 16 channels of video in a variety of video formats using spiking neural networks rather than convolutional neural networks (CNNs). The BrainChip Accelerator card is based on a 6-core implementation BrainChip’s Spiking Neural Network (SNN) processor instantiated in an on-board Xilinx Kintex UltraScale FPGA.

Here’s a photo of the BrainChip Accelerator card:

Note, this implementation is based on spike-model, which is quite DIFFERENT from convolutional neural network, such as GoogleNet, Alexnet. It is also LESS popular. Qualcomm and IBM are used to be the active researcheres in spike-models.

The following image shows a comparison of the spike model and convolution neural networks:

This is the speed comparison:

Written on September 13, 2017