Deep Convolutional Neural Networks have become a Swiss knife in solving
critical artificial intelligence tasks. However, deploying deep CNN models for
latency-critical tasks remains to be challenging because of the complex nature
of CNNs. Recently, FPGA has become a favorable device to accelerate deep CNNs
thanks to its high parallel processing capability and energy efficiency. In
this work, we explore different fast convolution algorithms including Winograd
and Fast Fourier Transform (FFT), and find an optimal strategy to apply them
together on different types of convolutions. We also propose an optimization
scheme to exploit parallelism on novel CNN architectures such as Inception
modules in GoogLeNet. We implement a configurable IP-based face recognition
acceleration system based on FaceNet using High-Level Synthesis. Our
implementation on a Xilinx Ultrascale device achieves 3.75x latency speedup
compared to a high-end NVIDIA GPU and surpasses previous FPGA results
significantly.

本研究通过探索多种快速卷积算法，包括 Winograd 和 FFT，并发现了一种将它们应用于不同类型卷积的最佳策略；实现在基于高级综合的可配置 IP 人脸识别加速系统中使用 FaceNet，并利用并行化的优化方案在新型 CNN 体系结构上，实现比高端 NVIDIA GPU 快 3.75 倍的延迟加速，并显著超过先前的 FPGA 结果。